How
Will the Diagnostic Company Business Model Evolve to Incorporate
Pharmacogenomics
Click the title to purchase "Successful
Pharmacogenomics Business Models"
Click to view TOC "Successful
Pharmacogenomics Business Models"
Click to view: "New Study Addresses
Key Issues and Requirements Shaping the Adoption of Pharmacogenomics
Testing"
An Interview with Stephen Little of DxS Ltd.
Emerging pharmacogenomics technologies are presenting the drug discovery
and development community with the promise of targeted therapies and
potential new revenue streams. Pharmacogenomics is also creating new
opportunities for diagnostic companies to develop tests that can be used
to determine patient response to a drug and screen out inappropriate
candidates. The key—and challenge—to the widespread adoption of
pharmacogenomic testing will be to successfully navigate the evolving
regulatory requirements and commercial challenges to formulate business
models that will enable diagnostic companies to forge mutually
beneficial partnerships with drug companies.
In this article, Stephen Little, Chief Executive Officer of DxS Ltd.,
discusses the market opportunities, challenges, and outlook for emerging
pharmacogenomic tests and the possible business models for diagnostics
companies to use their tests in combination with therapeutics. A
complete analysis of the current and future outlook of pharmacogenomic
testing is presented in a new Cambridge Healthtech report, Successful
Pharmacogenomics Business Models.
What is the current status of pharmacogenomic testing in the
healthcare arena?
The great hope of pharmacogenomic testing is that its use will usher in
an era of personalized medicine where a diagnostic test will be used to
identify patients likely to respond well to a particular therapy. In
reality, there are very few actual examples of pharmacogenetic
diagnostics, but we are seeing extensive use of genetic analysis in drug
discovery and drug development. The idea of using a diagnostic
hand-in-hand with a therapy has been around for a long time, but I think
the level of activity we are seeing, coupled with a general expectation
within the industry, means that personalized medicine in just around the
corner.
In general, there is a lot of support from all of the players involved
in the development and implementation of personalized medicines. These
players are the patients, the doctors, the regulators, the payers, the
diagnostic companies, and the pharmaceutical companies. It’s easy to see
how the patients and the doctors benefit from diagnostics that are
closely linked to specific therapeutics, because it results in a better
chance of getting the right treatment for the patient the first time.
It’s easy to see why the regulators are enthusiastic about the pairing
of diagnostics with therapeutics because they will see if a drug meets
patient needs more accurately. Similarly, the payers are enthusiastic
because they are paying for a result. The diagnostic companies are very
interested in the idea of capturing some of the value of a therapeutic
through a diagnostic test. The only player that is not completely on
board yet is the pharmaceutical industry, because they are concerned
about changing their usual way of operating. Even though the drug
industry is very inventive in terms of finding new medicines, they are
also very conservative in terms of the way they want to go about selling
these new medicines.
Why are some pharmaceutical companies resisting pharmacogenomics?
You do hear objections from pharmaceutical companies that
pharmacogenomics will reduce their market sizes. But that’s because they
consider their market for a drug to be all people with a particular
disease, when they should consider it to be only those people who are
actually going to respond to their treatment. So with pharmacogenomics,
the market for a particular drug doesn’t change, it’s just becomes more
accurately defined. The objection is that if you combine diagnostics
with therapeutics, you stratify the market, but in reality the market is
normally already stratified. For example, AstraZeneca has just launched
a new statin, Crestor, into a market where Lipitor, Zocor, Pravachol,
Lescor and Mevachol are already well established competitor products.
I think that the major drug companies are naturally cautious when it
comes to developing and selling their products, and pharmacogenomic
testing can be viewed as an unwanted complication. It is easier for them
to see the difficulty of the diagnostic rather than the profit of
personalized medicine. This is where the regulatory agencies can help by
ensuring that the drug industry is given every encouragement to do the
right thing and target therapies more effectively.
Ultimately, I think we will see three factors which will drive
personalized medicine forward. The first is regulatory pressure. The
second is that we need one or two more good examples which demonstrate
that a targeted therapy can also be a profitable therapy. Certainly
current therapies such as Herceptin and Gleevec which have a companion
diagnostic are generating significant sales, but a few more successes
would be very helpful. There is no substitute for success in encouraging
a trend. It may well be the case that more innovative drug companies
push ahead with diagnostic/therapeutic combinations to get their drugs
to the people who need it. Finally, the diagnostic industry can work
independently of the pharmaceutical sector to bring pharmacogenomic
tests to market.
What opportunities does pharmacogenomics present for diagnostic
companies?
There are two distinct opportunities for using diagnostics in
combination with therapeutics. On the one hand, there are diagnostic
tests that will allow the selection of one of several therapies; on the
other, there is a diagnostic test that is a prerequisite to prescribing
a therapy. I think it’s important to understand the difference between
these two opportunities. In the first case, the diagnostic company is
offering its customers the prospect of an improved therapeutic response
by selecting the most appropriate drug from a range of several
possibilities. To go back to the situation with statins that I mentioned
earlier, you could imagine a test which would allow the doctor to choose
the best statin to suit his patient. In the second case, the diagnostic
is what I’d call a therapy authorization test, meaning the test meets a
drug label requirement to pre-identify individuals who should or should
not receive the therapy. I think therapy authorization tests will be
developed at the behest of the pharmaceutical industry whereas therapy
selection tests could be developed in concert with the pharmaceutical
industry or independently.
The business strategies for attacking these two models—therapy selection
and therapy authorization—will be different. Therapy selection tests
will probably be developed and sold independently of the pharmaceutical
industry. This is because they will not tend to favor any one company’s
drug over its competitors, so as far as pharma is concerned their
overall effect will be neutral. The tests will be sold as traditional
diagnostics and marketed as a way of improving therapy safety or
efficacy and the diagnostic industry will naturally attempt to maximize
the revenues available from these tests.
The situation is different when the drug industry needs therapy
authorization tests to allow them to sell their products. In this case,
the diagnostic is the gateway to the therapy and the drug industry will
have a great interest in ensuring that these tests are readily
available. Furthermore, drug companies will have a different view of
these tests than diagnostic companies. For them the priority is not to
maximize diagnostic revenue but rather to minimize the chance of the
diagnostic delaying the launch or hampering drug sales. The key issues
will be timing, availability, convenience for the doctor, and regulatory
compliance. If that means accepting a lower return on the diagnostic for
an increased drug sale that is exactly what will happen.
I expect that two different types of businesses will evolve: the first
will follow a traditional diagnostic business model; the other will be a
business that understands and meets the requirements of a drug company
eager to get a product into the marketplace. This second area is where
DxS is positioning itself as a business right now.
How do you see diagnostics improving drug sales?
There are three ways of looking at it. The most dramatic situation would
be if the regulatory authorities are proactive and demand diagnostic
tests for some indications. In those cases there couldn’t be any drug
sales without the appropriate test.
The second approach is to use the promise of improved patient response
as a marketing tool to sell the therapy to doctors and health
authorities. I have never worked as a drug rep, but I’m sure that if I
did I would prefer to visit my clients armed with data that shows how my
product will be safe and effective in the patients that take it.
Certainly from a patient perspective, if you have a drug that might work
and one you know will work, you know which one you would choose.
The third opportunity, and I don’t know if this will become big or not,
but as drugs come to the end of their patent life, it will be very
interesting to see if drug companies are able to get patent extensions
by using a drug/diagnostic combination to identify a new sort of
treatment, such as using the drug in a smaller set of patients, or using
a smaller dose. It seems like an idea that pharmaceutical companies
should adopt willingly. When a drug is coming to the end of its patent
life, the use of a diagnostic to sell that drug in a different way could
get them another 20 years out of it. It’s more difficult, but if you can
differentiate a branded product from a generic by whatever means, it
gives you an extra edge.
What do you think is the likelihood that regulatory agencies will
make pharmacogenomic tests a mandatory part of the drug approval
process?
I think it’s a very good likelihood. For example, over the last few
years, there’s been some interest in using a TPMT test to identify the
2% or so of patients who are poor thiopurine metabolizers. The benefit
of the test is very high because the dose can be dangerous to patients
who have a reduced tolerance for the thiopurine class of drugs. It will
be easier to make that kind of requirement for drugs before they reach
the market than for drugs that are already on the market because you can
deal with it in a more considered way upfront. But where there is a
safety issue, I would expect pharmacogenomic testing to be driven by the
regulatory authorities if the drug companies don’t choose to do it
themselves.
Do you see a scenario where failed drugs are out-licensed to a second
entity?
It seems that you ought to be able to do that, but failed drugs seem to
have such a stigma attached to them that no one wants to go near them
anymore. No one seems to be doing it, though it seems like a perfectly
reasonable thing to do.
What about a targeted therapy, where a specialty pharma or a large or
medium biotech is willing to accept something other than the blockbuster
revenue cycle because the drug has a higher likelihood of being approved
due to the diagnostic test?
That’s a good point. There are two reasons why drugs have to be
blockbusters; one is that it cost so much to develop them, and the
second is that it cost so much to sell them. If you have a diagnostic
that improves both the chances of getting a drug through development so
that its development costs are lower, and at the same time improves its
performance so that more doctors are willing to prescribe it, it may
well be the case that you don’t need to have such huge drug sales to
still have a very profitable product, particularly if the treatment
commands a premium price. So I think it’s absolutely reasonable that
we’ll see companies try to do that.
Will pharmacogenomics mean the end of blockbuster drugs? I doubt it.
Looking forward, I think that personalized medicines will still be the
exception rather than the rule and that there will still be a place for
the blockbuster. Some people suggest that there may be a blockbuster set
of treatments, so you might have a drug company selling a combination of
treatments for different flavors of the same disease.
How will the diagnostic company business model evolve to incorporate
pharmacogenomics?
In light of the comments I made earlier about the two different types of
diagnostic requirements, you have to be clear about what kind of company
you are and what kind of business model you will follow. Are you going
to be a diagnostic company that sells a test to the healthcare industry,
or are you going to be a diagnostic company that teams up with a
pharmaceutical company to get their products into the marketplace? There
are both feasible models but with quite different requirements. I think
the more traditional model favors the existing diagnostic industry with
its well-established sales force, installed instrument base, and the
financial resources to demonstrate the clinical utility of their
products. The partnership model is more suitable for a smaller company
with great technology and innovative ideas about how to deliver the
diagnostic in the right place at the right time and to make sure it is
very widely available, has regulatory approval, is very easy to use, and
very affordable for the end-user.
Actually for most pharmacogenetic tests, the opportunities aren’t huge
in terms of numbers of tests performed. The really big molecular
diagnostics are tests such as HIV and HCV that need to be performed on a
regular basis. Most personalized medicine tests only need to be done
once. The diagnostic industry is so interested in pharmacogenomics
because of the lure of drug money. The drug industry, on the other hand,
is not interested in the diagnostic industry capturing the value of
their products; they want to keep the value for themselves, so there’s a
conflict that is creating an opportunity for new business models to help
drug companies accomplish their goals without giving their value away.
There is a huge opportunity for pharmacogenomics in the future, but it’s
uncertain as to when it will occur. There are hardly any pharmacogenomic
tests being used in the clinic right now, so diagnostic companies need
to position themselves so that they can be active when the time comes,
but they also have to make money in the meantime. You can either go down
the marker discovery route, or you can position yourself as a provider
with an understanding of what you think the drug companies are going to
need when the time comes, which is exactly what DxS is doing right now.
Click to view TOC "Successful
Pharmacogenomics Business Models" Click the title to purchase "Successful
Pharmacogenomics Business Models"
|
