Please login to the form below

Not currently logged in

Slow progress

Though a promising sphere, personalised medicines development can be limited when the economics do not add up for the diagnostics partner

Tortoise walking along a road'Personalised medicine' is a term that has been in use for years, dating back to before the human genome mapping project first got underway in the early 1990s.

However, despite major advances in understanding of the fundamental biology and chemistry of diseases over the intervening 20 years, as well as a huge investment in technology and people by the pharmaceutical industry, only a handful of personalised medicines have reached the market, 'poster-child' drugs such as Roche's Herceptin (trastuzumab) for HER2-positive breast cancer notwithstanding.

"Developing personalised medicines is not an easy task," noted Cecilia Schott, director of business development for personalised healthcare at AstraZeneca (AZ), who chaired a conference on the topic in London, UK, in October 2010.

Aside from the scientific hurdles of finding and validating sufficiently predictive biomarkers, it requires close collaboration with so many players, including pharma, companion diagnostics/theranostics companies, contract research organisations, plus academic and medical institutions. This can pose challenges not only for managing a project effectively, but also allocating intellectual property rights, according to Schott.

Pharmaceutical R&D departments have had to migrate from processes based on the development of high-volume, blockbuster treatments to high-value targeted treatments. Initially, fear of change and the impact on established business models was a factor, although Herceptin's success and other examples, such as the epidermal growth factor (EGFr) inhibitors, have reassured drug makers that personalised medicines can bring in attractive sales volumes.

Towards the end of 2010, the US Tufts Center for the Study of Drug Development (CSDD) published the results of a survey of top pharmaceutical companies which estimated that 12-50 per cent of all drugs in current clinical pipelines involved some degree of personalisation, despite the fact that this was 'a disruptive technology which does not come either easily or cheaply'.

"Early indications show that development of personalised medicines is commanding more resources and fomenting more organisational change than is generally appreciated outside the industry," said Christopher-Paul Milne, the author of the Tufts study.

Some companies have clearly placed personalised medicine at the heart of their R&D strategy. AZ, for instance, now evaluates all its new drug projects to see if they are compatible with a personalised healthcare approach, according to Schott. In 2010, the drug maker entered into a wide-ranging collaboration with diagnostics group Dako to help achieve that, saying at the time that around 10 per cent of its pipeline projects were personalised.

Pharma's wholesale shift towards personalised medicine is unsurprising given that the case for bespoke treatment is compelling, at least from a public health perspective.

Adverse drug reactions have been reported to be the fourth-leading cause of death, costing the US alone an estimated $177bn a year. With the typical efficacy rate of medicines across populations only around 50 per cent, an estimated $300bn out of the global $600bn medicines spend is at best providing no more than a placebo effect, according to Iain Miller, executive director of theranostics at in vitro diagnostics (IVD) company bioMerieux.

While pharma seems to be forging ahead, however, the IVD sector has been, to some extent, 'disincentivised' from participating in that effort, in Miller's view. He noted: "Deal flow remains modest and flat. There were more than 2,000 compounds in development but only seven IVD partnerships in 2008, even though around 10 per cent of those compounds make use of laboratory biomarkers."

Many of the drugs in development were in early-stage development, said the Tufts report, and that may explain in part why the number of deals involving companion diagnostics remains on the low side.

More serious, however, are the financial and value disincentives for the diagnostic partner, according to Miller.

"Our traditional products are high-volume, low-pricing tests, but companion diagnostics are low-volume, low-pricing. Few IVD products are based above the hundred dollar mark, so when patient populations are in the low thousands, the numbers are pretty modest."

Giving an example, Miller pointed to one of the success stories in personalised medicine, namely the EGFr inhibitors. Eli Lilly and Bristol-Myers Squibb's Erbitux (cetuximab), Roche's Tarceva (erlotinib) and Amgen's Vectibix (panitumumab) have combined sales in billions of dollars a year and the availability of diagnostic testing to identify non-responders is estimated to save the US healthcare system alone around $600m. For comparison, the market for IVDs comes in at just $10-$20m.

"This model fails from the perspective of aligned incentives ... even though it is a very exciting clinical story," maintained Miller.

That low return, coupled with increased risk compared to traditional IVD development, makes it hard to form a business case internally for companion diagnostics, and Miller believes the answer is for pharma to partner and help fund the initial R&D into companion diagnostics in order to make them more realistic value propositions for the diagnostics partner.

A business model is needed which incorporates elements such as the pharma partner funding the assay development cost, typically in the order of a few million dollars, and a modest success fee after significant drug sales have been demonstrated.

There have also been examples of pharma companies funding testing, such as AZ's sponsorship of EGFr mutation testing in the UK, which can make sense when access to the test is acting as a barrier to treatment with the drug.

One project which involves close collaboration between the pharma and diagnostics partner is Pfizer's non-small cell lung cancer candidate crizotinib (PF-0234 1066), which inhibits the anaplastic lymphoma kinase (ALK) oncogene.

Crizotinib is something of a pathfinder in personalised medicine because the companion diagnostic has been co-developed with the drug from the outset, rather then being added in later. "Typically, clinical teams get beyond phase I and then start looking around for a diagnostics partner, which can lead to delays," said Hakan Sakul, senior director in Pfizer's translational oncology group.

Pfizer's diagnostics partner is Abbott Molecular, and the company has developed a fluorescence in situ hybridisation (FISH) assay to test for activation of the oncogene which will be submitted for approval in the US and Europe at the same time as the drug.

Financial details of the partnership have not been divulged, so it remains to be seen whether that close collaboration will yield compelling results for the diagnostics partner. A rolling regulatory submission for crizotinib as a second-line treatment in non-small cell lung carcinoma (NSCLC) has just begun in the US.

Niko Drivas, senior licensing manager at Abbott Molecular, believes personalised medicine needs another large success story like Herceptin and the EGFr inhibitors to lend some additional momentum to the field.

"It's time for new companion diagnostics and new drugs, and oncology is ripe for personalised medicine because of the low efficacy rates and emerging science of what drives the tumour," he said.

Drivas agreed that the low return on investment in personalised medicine was a concern for diagnostics firms, given that even the most successful diagnostic is unlikely to make more than around $50m, often a fraction of what is brought in by the therapeutic component.

But he thought  it was important to look at companion diagnostics as a portfolio investment, and not look at the dynamics of just having one on the market. "We believe that when we have multiple deals in place for companion diagnostics the numbers will start to look very good," he commented.

There is also something of a chicken-and-egg situation between diagnostics and pharmaceutical firms, with pharma preferring to see the diagnostic available months before drug launch - ideally so it could be used in clinical trials - and the diagnostics partner wanting to see the therapy approved and successful before rolling out the test.

"Mitigating the risk and capturing value for both parties becomes important," according to Drivas. One way around that could be to launch a diagnostic with an analytical claim first, rather than one linked to treatment with a drug.

"The bottom line is that the IVD industry is the only industry that can bring you global, standardised tests that will allow personalised medicine to take place," noted Miller.

"If it doesn't work for us, it doesn't work."

The Author
Phil Taylor
is a freelance journalist specialising in the pharmaceutical industry

To comment on this article, email

6th April 2011


Subscribe to our email news alerts

Featured jobs


Add my company
Evoke Kyne

Evoke Kyne is an award-winning communications agency that harnesses insight, experience and creativity to advance health and wellbeing around the world. We were founded on the belief that...

Latest intelligence

Partnering with Patients and Caregivers in Clinical Research
Practical advice for Early Scientific Advice (ESA) in HTA submissions
A step-by step guide on the benefits, processes, and key considerations involved in the Healthcare Technology Assessment (HTA) Early Scientific Advice (ESA) consultation as part of planning the holistic evidence...
Jonathan Cooke
The fight against antimicrobial resistance
By Jonathan Cooke...