The Cross-Border Biotech Blog

Biotechnology, Health and Business in Canada, the United States and Worldwide

Tag Archives: Personalized Effectiveness

Biotech Trends in 2011: Comparative Effectiveness and Personalized Medicine

When this blog was launched in 2009, comparative effectiveness and personalized medicine were fairly new features in the North American landscape. Our initial argument that they were related topics — determining which treatment is best depends on which patient is being treated — was soon bolstered by the comparative effectiveness provisions in the U.S. stimulus bill and new personalized medicine data via the FDA.

The proposition has since become common knowledge, culminating in statements by Francis Collins and at BIO 2010 and discussed in the New York Times. Personalized medicine is now a key strategy for 12-50% of current drug pipelines, according to a recent Tufts study, and is a significant driver for DNA sequencing technology companies. If anything, the pendulum has swung a bit too far towards ‘hype,’ and as Matthew Herper reminds us, there are still non-personalized potential blockbusters in the pipeline.

The two concepts have even merged in a motto:

“the right drug to the right patient at the right time,”

which I still don’t like as much as “Personalized Effectiveness” (my neologized mash-up), but seems to be sticking. We’re just going to call it “Personalized Medicine” for now and will continue to follow major developments. You can too, on this page.

This post is the second in a series briefly outlining the biotech industry trends we’ve been following on the blog and noting some recent developments, plus directions for 2011.

Biotech Trends Update — Personalized Medicine: Duncan’s Personalized Health Manifesto is Primarily Preventative

Image from flickr user Steve Rhodes. Some rights reserved.Journalist David Ewing Duncan’sPersonalized Health Manifesto” was published this week by the Ewing Marion Kauffman Foundation. The most interesting thing about the manifesto* is that it assumes that the technical hurdles to generating and understading a full set of personalized health data have been overcome, and focuses on how that information can be deployed most effectively.  Duncan says that “a widening gap exists in integrating and implementing this promising new epoch of personalized health.” There are two main themes in the manifesto: integration and prevention.

Integration comes from Duncan’s view about how the flood of personalized data should be analyzed by researchers and physicians:

“A balance between specialization and integration needs to be restored,with an emphasis on the whole human organism as much as its parts…”

Prevention is, for Duncan, the natural best use of personalized data:

“Shifting to a health care system based as much on healthy wellness as illness is achievable…”

Both are laudable long-term goals, but I am not convinced of the need for urgent action on either point.

Specialization is (as Duncan acknowledges) what has enabled us to discover personalized markers and to analyze them on a allele-by-allele basis. We are a long way from making meaningful predictions about systemic effects on complex traits based on the available information. A shift too early away from specialization could prevent us from ever developing the underlying science sufficiently to make reliable predictions.

And shifting to a preventative healthcare system is not a goal I view as being unique to personalized medicine. As genomic testing becomes widely available, and patients begin to process their data, complex traits continue to present a challenge for them and their doctors — now that I know I have an increased risk of heart disease, I should shift my diet and increase my exercise. But these are things preventative healthcare advocates have been recommending for decades; and there is no evidence that I’ve seen that suggests the genetic information about increased risk is more motivational than family history, or peer behaviour, or any other non-personalized factor.

Bottom line: I’m no expert on manifestos, but while there’s nothing in this one I feel strongly opposed to, it doesn’t move me to action either. Read the whole thing (it’s not very long) and form your own (personalized) view.

* Other than his decision to call it a manifesto.
Image from flickr user  Steve RhodesSome rights reserved.

BIO Panel on Comparative Effectiveness Research Notes “Silver Lining” of Personalized Medicine

Speakers Daniel Todd, from EMD Serono, and Steve LaPierre, from Boston Scientific, were led by Foley Hoag lawyer Jayson Slotnik in a discussion of the final CER legislation and predictions about implementation. The overall tone was skeptical — the panel noted the potential for CER data to ultimately contribute to CMS coverage decisions, and worried about the cost of prospective randomized trials and about potential impacts on the FDA approval process.

They were, however, optimistic about the role of personalized medicine in the CER implementation. Steve LaPierre expects it to be helpful, and Daniel Todd advocates using the reference to molecular and genetic subtypes in the legislation to push for personalized analysis if a product is selected for a CER study. He calls personalized medicine a silver lining.

The panel also noted positive structural aspects, including helpful governance provisions in the CER legislation. Specifically, they were impressed with the public reporting and audit provisions and the availability of comment periods to allow private sector input.

In looking at how CER will shake out, the panel expected progressive adaptation of the program over the next 3-7 years.

Daniel Todd emphasized the importance of picking a first recommendation to establish credibility. The controversy this year about breast cancer screening recommendations shows the loss of credibility that can come from a debatable result, so they expect CER to ramp up over time.

He also predicted that dissemination of CER results through social media may drive bottom-up adoption of findings, and that defensive medicine may also contribute to adoption.

Bookmark and Share

Comparative Effectiveness and Personalized Medicine are “Part of the Same Question” Collins Confirms

In a very informative Kaiser Health News interview (via GenomeWeb), Francis Collins says that

“personalized medicine strategy and CER strategy are part of the same question. … There will often be more than one therapeutic intervention, so you have to compare them. But you also want to know what’s different about the individual that might have an influence on that answer.”

Couldn’t have said it better myself.

Bookmark and Share

Biotech Trends Update: A Personalized Critique of Comparative Effectiveness Misses the Mark

As the U.S. and Canada move to invest and rely more on comparative effectiveness research (CER), lack of personalization has been the loudest and most frequent objection.  That is why we have been following the interaction between comparative effectiveness and personalized medicine as a key industry trend.

Yesterday, an opinion piece in the WSJ by Leonard Zwelling, a professor of medicine and pharmacology at M.D. Anderson, came out strongly in opposition to CER.  Zwelling takes a number of swipes at CER, but most of the time he is focused on the inability of CER to personalize treatment — for a patient’s genes, age or preferences — in identifying a “best” treatment.  For example:

“If decisions based on CER inhibit the progress of personalized medicine—or in any way interfere with a meaningful interaction between doctor and patient to individually tailor the most appropriate therapy—no one is helped.”

Zwelling’s argument sets up more than a few straw men — that CER uses only retrospective data, that it will ignore qualify of life, and that treatment options change too quickly for CER to provide timely advice — but the main problem is that he assumes that CER and personalization are incompatible.

As this blog has noted on many occasions, CER is at its best when coupled with personalized medicine.  This post about KRAS genotyping is a great example.  This point is not lost on the Obama administration, which is aware of and sensitive to the need to account for personalized treatment in CER.  Zwelling himself opens with a quote from relevant legislation that says the current CER funding is supposed to find out

“what works best for which patient under what circumstances.”

The clause “for which patient” is a built-in acknowledgement of the importance of personalized approaches to CER. 

My bottom line: Both comparative effectiveness and personalized medicine are critical to medically effective and financially sustainable medical care and drug development.

Biotech Trends Update: Costs Savings from Personalized Medicine Sought by PBMs, Employers, Pharma Face Legal and Privacy Hurdles

When AstraZeneca announced a companion diagnostics collaboration recently, their head of oncology development said the goal was to get “the right treatment, to the right patient, the first time,” a nice turn of phrase* that is becoming a chorus in the healthcare industry.

This week, giant PBM Medco purchased DNA Direct, saying “[o]ur whole thing at Medco is to get people on the right drug the first time.”  DNA Direct uses its research on 2,000 available tests to help physicians, health insurance companies and patients understand how to use personalized medicine.  This is a good move — we said last month that education is key to expanding the personalized medicine market

AstraZeneca, Medco and other providers, employers and insurers would all like to use information on individuals’ health risks in order to reduce their costs, and as the Wall Street Journal reports, they are willing to provide incentives to their employees to mitigate those risks.  However, some of these efforts conflict with barriers put in place by the Genetic Information Nondiscrimination Act (GINA), which prohibits the intentional acquisition of genetic information about applicants and employees, and imposes strict confidentiality requirements on data that is acquired. (H/T @genomicslawyer)

In addition to legal barriers (some still being erected), AMA and other advocacy groups have also reportedly expressed concern.  I agree there is risk inherent in putting the decision of what the “right drug” is in the hands of manufacturers or payors, neither of whom is neutral in the outcome.  Medco, in particular, does not seem a neutral player here (at least based on their approach to Plavix and Effient, though I invite comments if I’m misinterpreting that study).

Still, a solution is required.  As I have been saying for over a year, personalized approaches to treatment have the potential to benefit all participants in the healthcare system, as the KRAS-Erbitux story has proven.  As Procter & Gamble said when investing in Navigenics’ funding round this week, “Personalized genetic testing can have significant meaning in helping consumers focused on prevention and wellness live better, healthier lives.” 

My bottom line:  A large part of the problem here is the low level of trust from the public, which even limits governments’ ability to act.  That’s particularly unfortunate, because government is the closest thing we have to a neutral funding source for comparative effectiveness and personalized medicine research (despite also being a payor). This is a problem much bigger than just personalized medicine, but until trust is restored, valuable cost savings and health benefits will go unrealized.

Bookmark and Share

* A concept I’ve been trying to call “personalized effectiveness” — tell your friends.

Top Four Biotech Trends of 2009

These may not all be consensus picks (and don’t miss the IVB’s year-end deal-centric fun) but I’m sticking with these four trends as the ones that have really shaped the year that was:

  1. Follow-on Biologics. Call them what you want (we like “biosimilars”, but we’re internationalist like that), there’s no denying that biosimilars were a major force in the industry and in politics this year.  Some in the press are calling the 12-year exclusivity period a done deal, but I say don’t count your chickens ’til the fat lady sings.
  2. Comparative Effectiveness and Personalized Medicine (not a two-fer, a trend of convergence). How much of comparative effectiveness variation will turn out to be an artifact of genetic sub-populations (each with binary responses to the drugs in question)?  Nobody knows, but as money pours into both fields, the truth will set us (and drug pricing) free.
  3. Shifting IP Constituencies. What do you get when you cross generics-hungry pharma companies with innovation-hungry Asian countries?  A whole new world of collaboration that will ultimately change the face of TRIPS and pharma R&D.
  4. Electronic Medical Records. Not really biotech (if you want to be picky), but it will have a massive impact on the way physicians, patients and payors interact with each other and with drug companies over the coming years.  Plus, with billions allocated to electronic medical records in Canada and the U.S., the pace of innovation and implementation really took off.

Bookmark and Share

Preventing Bias in Comparative Effectiveness Research

Comparative effectiveness research has the potential to avoid wasteful spending and create net benefits for patients if approached properly, but it’s expensive.  Many of the large-scale comparative effectiveness studies include industry funding, and benefits managers are no strangers to the game, but giving those partners a say in study design risks introducing bias. 

An interesting example comes from today’s report that pharmacy benefits giant Medco is planning a head-to-head study of nearly-off-patent Plavix versus brand-new Effient.  The interesting tweak here is that the study will exclude people with a genetic variant (of the CYP2C19 polymorphism) who can’t metabolize Plavix.

This seems like another great example of personalized medicine informing a comparative effectiveness decision.  But, as the In Vivo Blog pointed out in an August post about Plavix and Effient, the effect of the CYP2C19 polymorphism on Effient efficacy is unknown.

So the PBM, with cost-saving incentives, is setting up a study to make payment decisions in which the efficacy of the (cheap) generic is boosted by excluding patients with the CYP2C19 polymorphism, with the validity of the comparison based on the untested assumption that there is no systematic bias to the branded product’s efficacy in the excluded population.  Am I missing something here?

The moral of the story: fund comparative effectiveness research through neutral parties and keep a careful eye on genetic and phenotypic subgroups to maximize the value of these important studies.

Bookmark and Share

Trends Update — Personalized Medicine: Montreal CRO ethica Licenses Artificial Intelligence Data Analysis Product for Stratification

B&W_DNA_sequenceethica Clinical Research acquired a worldwide exclusive license to Matrix Pharma’s  artificial intelligence (AI) data analysis platform.  Neither the form of consideration nor payment structure (up-front vs royalty etc.) was disclosed, but the deal is “valued at CAD1.25 Million.”  The companies say the AI can:

“extract interdependencies, correlations, and predictive models from complex data sets that conventional statistical tools are unable to detect.”

In other words, it’s an approach designed for stratification of clinical trial data to identify personalized subgroups for whom a drug may be particularly effective.

ethica says the product, which will be marketed as “eidyia™” (pronounced “idea”)

“has already yielded successful results in a series of clinical applications such as identifying biomarkers, optimizing the predictive power of biomarkers, identifying responder/non-responder indices, genotype-phenotype linkages, elaborating models for early diagnosis, and for classifying participants in Phase II clinical trials for the optimization of Phase III clinical trials.”

I don’t know if  “eidyia” is (or isn’t) the best platform for these tasks; but I am convinced that in many cases, the ability to stratify patient populations will be key to demonstrating the effectiveness of new drugs and winning approval and reimbursement.

Bookmark and Share

Trends Update — Personalized Medicine: Merck Strategy Head Skeptical

As I’ve been following personalized medicine on this blog, I have become almost convinced that recent advances in genomics technology put us at the brink of an era of personalized diagnosis and treatment.  Not everyone agrees.

Chris Morrison, reporting from the Pharmaceutical Strategic Alliances meeting, quotes Merv Turner (the head of strategy at Merck) as follows:

“‘You can reduce cardiovascular mortality by 50%’ by using statins, he said. ‘That means 50% cardiovascular disease is unsatisfied. Is that 50 different small diseases or one large one? Personalized medicine is like soccer in the US: it’s the game of the future and always will be.'” (emphasis added)

Maybe I’m missing some context here, but I think the answer to Turner’s question (50 different small diseases or one large one) is “we don’t know.”  It is an interesting question and I think it would be prudent to find out the answer.  Preferably before we try treating the 48th disease with a drug developed for the 35th.

Trends Update — Comparative Effectiveness and Personalized Medicine: Is Canada Ahead of the U.S. In the Use of HER2 Testing for Personalized Breast Cancer Treatment?

B&W_DNA_sequenceFor the 20%-30% of breast cancer patients with tumors that overexpress HER2, treatment with Herceptin (an antibody drug from GenetechRoche) is highly effective.  That’s why this article in the journal Cancer is so shocking.  The authors gathered data from a variety of published sources and estimate that:

“up to 66% of eligible patients had no documentation of testing in claims records, up to 20% of patients receiving trastuzumab were not tested or had no documentation of a positive test, and 20% of HER2 results may be incorrect.”

I asked a friend, who is a genetic counsellor in Toronto, if she thought the gaps here were as bad.  She said no, and that she would be extremely surprised not to see a HER2 result in a patient’s file.

Of course it’s not valid to draw conclusions about national health care from a comparison of the Cancer study to my anecdote.  Among other reasons, the authors admit their data is flawed and dated, and my friend works at one of Canada’s top teaching hospitals.  Nevertheless, the possibility that many U.S. patients could be falling through the cracks is extremely disturbing.

If clinicians can’t adopt and maintain near-universal use of a personalized medicine approach like HER2-Herceptin with recognized and measurable benefits, the future of pharmacogenomics is in some big trouble and we will never generate truly useful comparative effectiveness data.

H/T @FiercePharma.

Bookmark and Share

Trends Update — Comparative Effectiveness: Where Data Shows No Difference, Tie Should Go To the Patient

A post by Scott Hensley on the NPR Health Blog yesterday has some good food for thought in the comparative effectiveness debate: what to do when comparative effectiveness studies show no statistically significant difference between treatments.

The post notes that insurance coverage will be a factor in these decisions, but that:

“in the end, it might be you and your gut feeling.”

In one of this blog’s prior posts, I noted that it will be hard to distinguish between treatments that show different effectiveness because of personal differences between patients and those that would show different results even if the patients were identical. 

Hensley’s post illustrates that no matter how much data we gather, there will be gray areas where doctors and patients have to make subjective calls.  I hope payors will be extremely cautious about second-guessing these decisions.  Tie goes to the patient.

Bookmark and Share

Trends Update — “Personalized Effectiveness”: Amgen Gets Prospective Data to Back KRAS-Vectibix Plan

B&W_DNA_sequenceA few weeks ago, when the FDA changed the labeling on anti-EGFR drugs, Amgen was pretty enthusiastic about “avoiding unnecessary treatments in patients [with a specific genetic marker] who are unlikely to benefit” from Vectibix.  Avoiding these patients leaves more reimbursement available for patients who would benefit from Amgen’s product.

Now Amgen has even better data to support its personalized approach to colorectal cancer treatment: their study of Vectibix as a first-line treatment tracked the KRAS genetic status of participants and showed “significantly prolonged progression-free survival” for the wt-KRAS group.

In patients with mutated KRAS, Vectibix wasn’t just “unnecessary,” it actually showed worse outcomes than the control group, meaning genetic testing of all colorectal cancer patients will be a top priority.

A second important note for companies thinking about companion diagnostics and personalized effectiveness is that the Amgen study was a prospective study that will support much more robust conclusions.  H/T @ldtimmerman.

In contrast, recall that the CMS decision not to reimburse genetic testing for Warfarin dosing specifically cited the lack of prospective data on which to base a decision.

This new data from Amgen will:

  1. Drive tumor genotyping as a standard of care; and
  2. Help make the economic case for companion diagnostics.

Share Button

Follow

Get every new post delivered to your Inbox.

Join 130 other followers