Leaders in Pharmaceutical Business Intelligence (LPBI) Group

Diagnostics Industry and Drug Development in the Genomics Era: Mid 80s to Present

Author and Curator: Larry H Bernstein, MD, FCAP

This article is a Chronological continuation to my article

Milestones in the Evolution of Diagnostics in the US HealthCare System: 1920s to Pre-Genomics

http://pharmaceuticalintelligence.com/2014/11/16/milestones-in-the-evolution-of-diagnostics-in-the-us-healthcare-system-1920s-to-pre-genomics/

Diagnostics Industry and Drug Development in the Genomics Era: Mid 80s to Present

Part 1.1: Companion Diagnostics

http://pharmaceuticalintelligence.com/?s=Companion+Diagnostics

As the life sciences industry adjusts to the insurance-driven environment,

• specialized medicine is emerging as a critical asset in a development strategy.

Biomarkers and companion diagnostics (CDx) offer companies

• the ability to target a subset of patients for testing with some potential for greater efficacy
• based on biomarker selectivity, accelerated regulatory approval and better reimbursement.

Companion diagnostics are not necessarily suitable for developing a product. The decision to embark on a CDx program is a complex one involving a pharmaceutical company’s resources and its clinical goals. Top-performing companies prepare for success by

• developing consistent criteria and protocols that lead to thoughtful partner selection and follow-through —
• ensuring a smooth co-development timeline under budget.

Partner selection implies a cooperative development with

• a diagnostics company having the needed diagnostics selectivity.

Consistent criteria and protocols requires that

• the test criteria and performance characteristics accurately allow for
• objective measurement of disease progression or effective treatment.

Criteria to Select the Right Test for a Companion Diagnostics Program

3.2  Personalized Medicine: Why We Are So Excited

http://www.genengnews.com/bioperspectives/personalized-medicine-
why-we-are-so-excited/4767

In this first chapter of a five-part series, learn about new scientific developments
and recent treatment successes.

Detlef Niese, M.D., Ph.D.

Same Symptoms: Same Disease

Patients diagnosed with the same disease may react very differently to the same treatment. This observation led   Sir William Osler (1849–1919) to the statement: “If it would not be about the variability among individuals
medicine could well be a science and not an art”. We are brought up with the idea
that

• patients who have the same symptoms and whose course of disease follows similar patterns
• are likely to suffer from the same disease.

Scientific disciplines such as anatomy, developmental anatomy, physiology and
biochemistry,

• helped us to get a better understanding of how a normal organism functions.

Pathophysiology, micro and macro pathology and clinical chemistry

• helped to describe, classify and understand diseases.

However, few of these scientific advances led to a

• full understanding of the causes of diseases on a molecular, cellular or genetic basis.

Today’s medical text books still describe and define diseases on the basis of
combinations of objective and subjective symptoms, macroscopic, microscopic,
pathophysiological or biochemical findings, and the presence or absence of
specific pathogenic organisms or substances. Most diseases are also classified
according to the affected organ or organ system.

In consequence, patients diagnosed with the same disease

• are considered to suffer from the same underlying pathology.

They are also expected to respond to the same treatments, although clinical
experience tells us that this is only true in a limited number of situations. What is
causing the symptoms on a molecular level, remains often unclear.

Effective Therapies Were Educated Guesses

We have to admit that our current knowledge of disease mechanisms unfortunately
has not always resulted in the development of highly effective medicines yet.
The effective therapies we have, were developed intuitively or empirically, or were
even discovered by chance. Examples include antibiotics, alkaloids, calcium channel blockers, immunosuppressants and cytotoxic agents.

There are many examples of medicines, which were developed on the basis of wrong assumptions, and some examples of medicines, which proved to be effective in an
unexpected condition. But there are also few examples of medicines, like oral
contraceptives or modern targeted medicines in oncology and rheumatology, of which the development is based on an exact understanding of the relevant mechanisms.

Different Symptoms: Same Underlying Disease

In reality patients diagnosed with the same disease based on the same diagnostic
criteria

• show very different courses of the disease, and
• respond very differently to the same treatment. In clinical practice,
• doctors have no other choice than to optimize the treatment by trial and error.

We now know that these patients are probably suffering from different diseases.
In other words:

• the diseases have a different underlying pathology.

While the scientific progress in basic scientific disciplines such as anatomy,
physiology and biochemistry

• has been substantial over the last centuries,
• we still have a very limited understanding of the molecular processes in diseases.
• we know even less about the factors, which may determine whether a person has an increased risk
• of developing a particular disease.

Is “type-2 diabetes” a single homogeneous disease, for instance?

New Understanding of Diseases

Over the last decades molecular research on sick and healthy cells

1. gave new insights in cellular signalling pathways: mechanisms involving proteins and nuclear receptors,
2. which are essential for the normal functioning of cells
3. but under specific circumstances may lead to diseases.

These discoveries have helped us to understand and classify diseases in a new way.

We have learned a lot from rare monogenetic disorders, which allow us to study the
impact of specific genetic mutations on the clinical signs and symptoms of the disease.

This research allows us to gain insight in the

• underlying genetic alterations and
• the consequences of the presence and functioning of critical proteins.

The answer to the question ‘what causes diseases’ is revealing itself little by little

• since we are able to decipher the genetic code of a person.

3.3 Personalized Medicine: Translation of Concepts

http://www.genengnews.com/bioperspectives/personalized-medicine-translation-of-concepts/4783

This second chapter of a five-part series shows how the search for predictive biomarkers could change drug
development research.

Lasse Tengbjerg Hansen  ,    Adam Heathfield, Ph.D.
Biomarker research activities should continue

• in parallel with the clinical development program,
• so that biomarkers can be improved by applying data
• derived from early clinical trials to later stages of the program. [© taraki – Fotolia.com]

The previous chapter discussed why scientific developments and recent treatment
successes have made many of us so excited about personalised medicine. This
chapter describes how

• the search for predictive biomarkers is likely to change drug development research.

It may lead to a reduction of costs and an increase in productivity of research &
development projects.

Predictive biomarkers will change the way clinical trials are performed.

We Need Predictive Biomarkers

Traditionally, biomarkers answer

• critical questions that arise during various stages of drug development— questions such as:

1. Does the drug reach the target?
2. Does it have the desired biological effect?
3. Does it have an influence on other expected or unexpected targets?
4. Does the drug affect characteristics that predict desired or undesired effects?

However, the concept of personalised medicine centres around predictive:

1. biomarkers that can help select a patient population
2. with a higher chance for a favourable response
3. to a specific kind of medicine.

In today’s drug development programmes, the identification and qualification of
biomarkers are integrated.

If we are able to determine the

• biologically relevant dose, range and selection of the optimal target population by means of biomarkers,
• we are convinced that biomarkers will increase R&D productivity by reducing development timelines and
• will prevent costly late stage attrition.

Unfortunately, predictive biomarkers are often not applied until rather

• late in the clinical development programme, when clinical data show
• that an optimal benefit-risk profile is only achieved in a subpopulation of patients.

In such situations,

• the attention is suddenly turned to other available research data
• that might explain the underlying biological nature of the research results.

Over the last years we gathered examples of biomarkers, which

1. have proven to predict clinical response better
2. because they are linked to the mode of actions of the compounds in question.

Examples include Her-2/neu (Trastuzumab/Lapatinib), KRAS (Cetuximab/Panitumomab), BRAF
(Vemurafenib), and CCR5 (Maraviroc).

We Lack Model Systems to Predict Drug Response

While it is crucial to identify and qualify biomarkers that predict clinical response

• we lack good model systems that predict drug response.

Since efficacy is established in animal models, despite convincing data from animal models,

• the translation from animals to humans is not always successful.
• alternative models or an alternative technology are needed, and
• this had been attempted with ex-vivo systems in areas such as rheumatoid arthritis and Crohn’s disease.

These assays have proven to be valuable for the evaluation of novel therapeutic
targets.While the application of such assays seems promising, it remains to be fully investigated.

A caveat of ex-ivo systems may be that

• the generated biomarker signal derives from a local tissue environment, which
• might be difficult to capture in peripheral blood samples.

The oncology field has the best access to tissue biopsies. That is one of the reasons why oncology is a front runner in personalized medicine with several tissue based
companion diagnostics. On the other hand, still very little is known

• about the drug resistance as encountered in the oncology field.

This asks for biomarkers that elucidate more clearly what is going on in individual
tumors.

New Kind of Trials

Studying targeted medicines in traditionally defined patient populations is quite problematic. For example, if we want to test a medicine like gefitinib,

• we do not need patients who are clinically diagnosed with non-small cell lung cancer,
• but patients with a specific EGFR mutation.

If we were to test the medicine on a population which is

• not preselected for the presence of the mechanism in question,
• we might conclude that the medicine does not have an appropriate benefit-risk ratio.
• but if tested on patients with the underlying mechanism we might come to the opposite conclusion.

Gefitinib is therefore indicated for patients with tumors showing specific EGFR
mutations, while

• it is irrelevant whether the tumour is located in the lungs.

In order to develop targeted medicines, patients will have to be selected

• based on presence or absence of specific biomarkers.

In some situations, that may be a

• relatively small subset of the traditionally defined patient population.

For instance: traditionally we would test a drug on patients with breast cancer.
But now we would select patients with a

• similar disease pathway: e.g. Her-2 positive.

Eventually, the patient population may be extended: for instance –

• that we can treat patients with malignant tumours in other organs with the same medicine,
• assuming the tumours are caused by the same molecular mechanisms (e.g. m-TOR mutations).

With this approach, patient populations participating in clinical trials would be

• more homogenous, and response to treatment will be more consistent and predictable as well.

Integrating Biomarker Research with Target Evaluation

We think continuous biomarker research

1. ideally starts at least two to four years prior to first-in-man clinical trials,
2. we should ideally have identified a broad panel of biomarker candidates that
3. subsequently can be further qualified in appropriate in-vitro, ex-vivo, in-vivo, or in-silico model systems.

Bioanalytical assays for the selected candidates can then be established and
validated

• prior to implementation in clinical trials.

In order to be fully operational, this strategy requires competencies within three main areas:

1. biomarker research,
2. biomarker assay development and
3. clinical biomarker implementation.

These areas should collaborate to guarantee the quality of the biomarker data derived from clinical trials. This will strengthen the basis on which a

• personalised medicine program can be evaluated.
• it is also advisable to continue the biomarker research activities
• in parallel with the clinical development program.

In this way it is possible to improve the biomarkers by

• applying data derived from early clinical trials to later stages of the program.

This approach clinically validates pre-selected biomarker candidates

• but may also identify potential other and better biomarker candidates correlating with clinical outcomes.

Identifying Biomarkers is a Collaborative Effort

It is time consuming and costly to identify and qualify biomarkers for other uses,

• such as predicting clinical outcomes.

A biomarker needs sufficient evidence of broader clinical utility and validation

• in multiple independent studies,
• across different cohorts,
• ethnic groups and
• clinical subgroups.

This is in line with recent guidance for biomarker qualification from the Food
& Drug Administration (FDA) and the European Medicines Agency (EMA).

In recent years, large consortia have succeeded in

• the qualification and validation of various types of biomarkers.

Examples of collaborative biomarker onsortia are the Biomarker Consortium (US) and Innovative Medicines Initiative (EU).They are supported by representatives from
regulators, pharmaceutical industry, biotech industry and diagnostic industry,
academia and governmental organisations.

3.4 Personalized Medicine: From Biomarkers to Companion Diagnostics

This third chapter of a five-part series talks about coordinating diagnostic test kit development between
pharmaceutical companies and diagnostic manufacturers.

Peter Collins, Ph.D.

http://www.genengnews.com/gen-articles/personalized-medicine-from-biomarkers-to-companion-diagnostics/4820/

The previous chapter described how the search for predictive biomarkers

• is likely to change drug development research.

This chapter talks about turning biomarkers into companion diagnostics.

Once biomarker tests have been turned into diagnostic test kits, we would

• be able to routinely classify patients in clinics.

But the development of those diagnostic test kits has to be coordinated between two entities—pharmaceutical companies and diagnostic manufacturers—with rather different regulation.

As we have seen in the previous chapters, biomarkers can play a critical role in

• classifying patients into subpopulations.

A biomarker can be used as the basis for

• creating a routine diagnostic test for clinical use,

after it has been approved as an in-vitro diagnostic (IVD) test: i.e.

• a certified product, reagent, application or other tool that analyses human material and helps to diagnose patients.

There is a tendency for regulators to require that

• drugs and companion diagnostics are developed in tandem.

A companion diagnostic test is essentially a biomarker test

• that enables better decision making on the use of a therapy.

In other words: it is a diagnostic test that is

• specifically linked to a therapeutic drug.

The goal here is to increase the safety and the efficacy of the drug.

Pharmaceutical and biotechnology companies are trying to change their tack now and are working hard

• on integrating the co-development concept but
• true co-evelopment has been a rare phenomenon

There are two reasons for this.

1. clinically useful biomarkers are usually established late in the drug validation process.
2. the worlds of drug development and diagnostics, although both part of health care, are parallel universes

In the next paragraphs we will elaborate on these issues.

Clinically Useful Biomarkers Found Late in The Drug Validation Process

It is quite difficult to find clinically useful predictive biomarkers early on in a drug
development program, simply because

• they can only be determined on the basis of the patients’ responses to the drug.

A number of biomarkers, such as KRAS and EGFR mutations,

1. could only be established after a sufficient number of patients—
2. well beyond the usual number for a Phase III trial—
3. had elicited better understanding of differential drug response.

If the method of clinical trials changes consistent with our new understanding
of diseases and classifying patients (as we described in chapter one
and two), we might be able to find these biomarkers at an earlier stage. But

• as long as still 30% of the drugs fail during Phase III,
• diagnostic manufacturers will not be able to afford huge investments
•  in the development of companion diagnostics.

Parallel Universes

The worlds of drugs and diagnostics are parallel universes: in general

1. they have different development timelines, product lifecycles, return on investment, customers, and regulations.

• Drugs are valued and reimbursed as products, typically of high value.
• Diagnostics are valued and paid for as services,typically at a much lower value.
• Relatively few if any models exist for valuing a drug diagnostic combination.

This issue will be discussed in the next chapter.

Drugs are protected by patents, but in the companion diagnostic arena,

• biomarkers are considered to be within the public domain and there is less emphasis on intellectual property.

It is even debated whether biomarkers should be patented at all, because

• biomarkers are not invented but already exist in cells.

Drugs and diagnostics are also regulated differently. It takes many years and

• a large amount of data from expensive clinical trials to get a drug approved.

Under current rules, however in the current IVD Directive

• there is little mention of clinical evidence.

However, the EU IVD Directive regulating in-vitro diagnostics is under revision.
The current IVD classification

• does not take scientific and technological evolution into account
•  which is expected to be changed towards a new risk-based approach.

A revised IVD Directive (which may even become a Regulation)

• will improve safety and efficacy, but will also raise difficulties.

It is likely to state that an IVD assay must have the

• performance characteristics  required for fulfilling a clinical purpose.

Moreover, requirements may be added on

• how to demonstrate clinical validity, possibly proportionately with
• the risk level of the test.

The intention to increase safety will inevitably

• lead to higher costs and greater efforts for manufacturers.

It will take IVD manufacturers years and unusually high
investments to take an IVD past the regulators. Undoubtedly,

• Phase III trials would benefit from well validated biomarker tests.

But it is almost impossible to comply

• with regulations that require a clinically validated diagnostic test,
• simply because clinical evidence will only be provided
• by a clinical trial itself, and this is unlikely to be required for many IVDs.

Another point is that more regulation can severely impact innovation. The larger
investments that are needed might limit the ability

• of (especially small) innovative companies to discover and develop biomarkers.

The Quality of Diagnostic Tests

In order to gain more insight into the consequences of stricter regulation, we must

• dive into the universe of diagnostic practice in the EU a little deeper, because there is another challenging reality.

In the diagnostic arena there are, basically, two types of tests in use:

1. manufactured diagnostic products, which are regulated in the EU, and
2. lab developed tests (LDT’s), which are not regulated.

Thus, in the EU, there is a broad range of methodologies, often driven by

• the individual laboratory’s capabilities, access to diagnostic testing platforms,
  reimbursement and preferences of the individual team.

How this situation works in practice was pointed out by a multicenter study. Various laboratories were asked to select patients 

• with a positive EGFR or  KRAS biomarker.
• Some of the tests were in close agreement but, others were not.
• Due to the poor quality of so-called ‘home brew tests’ or inadequately validated tests in hospital laboratories,
• patients with a positive EGFR biomarker might be excluded from an effective therapy on false grounds.

The result of this industry dynamic is a lack of standardization of testing and,

1. inconsistent patient selection for therapy.
2. Improper patient selections will lead to poor therapy outcomes, or worse, to adverse reactions to drugs.

This situation is not likely to change under a new EU IVD Regulation. IVD
manufacturers may have to make greater efforts in the future, including financially, to get an IVD past the regulators. However,

• there is no legally enforced requirement
• to use the companion diagnostic test approved in the clinical trial.

So, for as long as regulatory bodies do not sanction laboratories that

1. do not conduct tests in line with pharmaceutical clinical trials,
2. the quality of the testing will not improve.

Diagnostic partners cannot justify large investments in bringing products onto a market, in which there are

• no rules for laboratories to perform substitute testing.

Regulation for testing  laboratories will be a more effective first step

• towards increasing quality and safety than sharpening the rules for IVD manufacturers.

Discussion

The industry needs time to bring

• diagnostic and therapeutic research together.

The regulation should leave room

• to develop companion diagnostics in various business models,
• instead of only requiring co-development.

Let’s not forget: many drugs

• have proven to be effective without companion diagnostic tests.

We would not want to rule those out.

Regulation to increase safety is in itself a good thing, but

• it is also important to provide opportunities for
• small innovative enterprises to enter the market.

3.5 BioPerspectives

Apr 3, 2013

Personalized Medicine: Health Economic Aspects

This fourth chapter of a five-part series discusses personalized medicine’s
financial impact.

Alexander Roediger

http://www.genengnews.com/bioperspectives/personalized-medicine-health-economic-aspects/4824

There is a widespread skepticism about the financial impact of personalized medicine. [AlienForce – Fotolia.com]

From the previous discussion of turning biomarkers into companion diagnostics,

• we now shift to a discussion of healthcare.

Personalized medicine has an impact on the healthcare budgets.

Are the tools that healthcare economists and pricing and reimbursement authorities
currently use

• suitable for analysing personalized medicine?

In short, the concept of personalized medicine revolves around the idea of the
selection of patients who are likely to

• respond the medicine, so that the treatment will offer better outcomes.

According to the recent Quintiles report “biopharma and managed care executives are

• optimistic that personalised medicine will improve efficacy, safety and public health”.

However, there is a widespread skepticism about the financial impact of personalized medicine. According to the same report

• 56% of managed care executives feel that personalized medicine will increase cost of prescription medicines.

Consequently, pricing and reimbursement of personalized medicines

• need careful consideration and a balanced view
• on cost-effectiveness and incentives innovation.

Another Economic Model is Needed

Much of our experience with economic evaluations of medicines is based on

• medicines designed to treat the whole patient population.

Evaluations of the economic value of a particular drug

• usually involve comparisons with treatment alternatives or palliative care.

Such comparative effectiveness are typically assessed

• on the basis of data collected in
• randomized controlled trials across a broad population of patients.

There have been attempts to

• identify sub-sets of patients that benefit most noticeably from the medicine, i.e.
  patients over or under a certain age or patients judged to have a ‘severe’ form of a
  certain disease.

But in many cases these evaluations were

1. not supported by robust evidence, and the sub-groups were not well characterized
2. nor were they statistically well-sampled in the clinical data.

Testing May Be Economically Viable

In the personalized medicine concept,

• a sub-group of responders is selected or screened out, which
• raises the hope that economic evaluations will be more straightforward and positive.

This is particularly true for the cheapest use of costly molecular targeted agents.

Biomarkers can improve the ability

• to identify responders and non-responders,
• and insure that the information is used to make better treatment choices.

As a consequence, the respective drug gains a  more favorable risk-benefit ratio,

• patients are expected to have better health outcomes and better quality of life and
• healthcare resources are likely to be used more efficiently.

Much of the efficiency gains of personalized medicines depends on the testing—

•  “testing before treating may be economically viable
• if the savings gained by avoiding ineffective treatment and adverse events are greater than the cost of testing.”

Studies on the cost-effectiveness of personalized medicines are still on the way
but promising results are available. The introduction of

• a companion diagnostic strategy in advanced non-small cell lung cancer
• reduced overall treatment costs by more than € 800 compared to current treatment.

A cost-effectiveness analysis in the field of chronic myeloid leukemia showed that

• the use of the FISH test reduced treatment costs by €12’500.6 In addition, it increased the life quality.

Reduction of R&D Costs

Personalized medicine may provide more value for money—

1. not only because of improved drug effectiveness and reduced toxicity, but
2. also decrease the average research and development costs for new medicines.

Clinical trials are the most expensive part of R&D (nearly 50% of the investment).
The costs of clinical trials seem

• to have risen by one third between 2005 and 2007 due to
• increasing regulatory and other requirements.

Biomarkers may enhance the efficacy of clinical trials of new drugs

• by investing more heavily in early research to identify key biomarkers,
• and in targeting relevant sub-groups of patients.

Smaller (and maybe even shorter) clinical trials are likely to reduce
development costs.

This sounds promising, but since personalized medicine is in its early stages,

• efficiency gains may occur only in the long run. Furthermore,
• companion diagnostics are likely to increase the additional costs and the complexity of
• the risky process of drug discovery and development.

In oncology,  where personalized medicine is currently progressing most rapidly,

• the late stage failure rate of new compounds has historically been higher than in any other area.

A well-known argument against personalized medicines is, that

• they will lead to smaller groups of eligible patients and
• therefore lead to higher unit prices in order to deliver competitive return on investment.

Additional value, therefore, “needs to be captured in terms of

• premium pricing, faster adoption or longer effective patent life for a portfolio of targeted drugs,
• to offset the reduction in potential revenues from market stratification”.

However, personalized medicines do not only diminish groups of eligible patients,

• they enlarge them as well (see chapter 2).

The following problems still have to be solved:

1. The costs of diagnostics are not easy to describe
2. The evaluation of cost-effectiveness is difficult
3. The regulatory pathway is fragmented

Costs of Diagnostics Are Not Easy to Describe

Implementing personalized medicines in healthcare is potentially a costly investment:

• it requires testing a whole patient population to identify groups of responding patients
• or to screen out patients likely to suffer adverse events or who need different dosing.

Some evaluations have attempted to tie the diagnostic

• to the use of a specific new medicine (“co-dependent technologies”) and
• have in some cases passed the costs of diagnosis on to the medicines developer.

But more and more multiple tests and multiple personalized medicines for particular
diseases become available. For example, in colorectal cancer and non-small cell
lung cancer,

• a set of parallel tests are to be performed on a number of molecular biomarkers
• to decide between a range of personalized medicines.

But more complex diagnostic and treatment pathways make it

• less easy to ascribe costs of diagnosis to the use of a particular medicine.

Evaluation of Cost-Effectiveness is Difficult

To evaluate the cost-effectiveness of particular molecular diagnostic approaches
is also problematic. Diagnostics are normally supported by

• analytical performance data and rarely by clinical or outcome data,
• but in order to perform a health technology assessment you need the latter.

According to a systematic review, only eight studies evaluated the clinical validity, andnone of the studies was a prospective evaluation of a test’s clinical utility.

In addition, personalized medicines may also require

• a different view on the clinical evidence available when a medicine is launched.

In some examples, personalized medicines have been approved on the basis of

• a retrospective analysis of clinical data identifying the responsive sub-population.

In other cases, personalised medicines have been launched under

• conditional approval mechanisms on the basis of phase 2 data alone (smaller studies, often
  without overall survival end points).

Regulatory Pathway is Fragmented

Discussion

Current health economic research shows that personalization of treatment, i.e.
identifying responders and non-responders, has

• the potential to improve effectiveness and reduce costs.

In addition, biomarker testing may lead to more successful R&D projects.
However, the value of so-called tailor-made therapies depends to a large
extent on the quality of the tailor. In other words,

• adaptations to both regulatory structures and market structures
• are necessary to encourage the development of personalized medicine.

Part 4: Biomarkers as Diagnostics

There has been consolidation going on for over a decade in both the pharmaceutical and in the diagnostics industry, and at the same time the page is being rewritten for health care delivery.  I shall try to work through a clear picture of these not coincidental events.

Key notables:

1. A growing segment of the US population is reaching Medicare age
2. There is also a large underserved population in both metropolitan and nonurban areas and a fragmentation of the middle class after a growth slowdown in the economy since the 2008 deep recession.
3. The deep recession affecting worldwide economies was only buffered by availability of oil or natural gas.
4. In addition, there was a self-destructive strategy to cut spending on national scales that withdrew the support that would bolster support for infrastrucrue renewl.
5. There has been a dramatic success in the clinical diagnostics industry, with a long history of being viewed as a loss leader, and this has been recently followed by the pharmaceutical industry faced with inability to introduce new products, leading to more competition in off-patent medications.
6. The introduction of the Accountable Care Act has opened the opportunities for improved care, despite political opposition, and has probably sustained opportunity in the healthcare market.

Let’s take a look at this three headed serpent. – Pharma, Diagnostics, New Entity
?  The patient  ?
?  Insurance    ?
?  Physician    ?

4 Biomarkers in Medical Practice

4.1 Case in Point 1.

A Solid Prognostic Biomarker

HDL-C: Target of Therapy or Fuggedaboutit?

Steven E. Nissen, MD, MACC, Peter Libby, MD

DisclosuresNovember 06, 2014

Steven E. Nissen, MD, MACC: I am Steve Nissen, chairman of the Department of Cardiovascular Medicine at the Cleveland Clinic. I am here with Dr Peter Libby, chief of cardiology at the Brigham and Women’s Hospital and professor of medicine at Harvard Medical School. We are going to discuss high-density lipoprotein cholesterol (HDL-C), a topic that has been very controversial recently. Peter, HDL-C has been a pretty good biomarker. The question is whether it is a good target.

Peter Libby, MD: Since the early days in Berkley, when they were doing ultracentrifugation, and when it was reinforced and put on the map by the Framingham Study,^[1] we have known that HDL-C is an extremely good biomarker of prospective cardiovascular risk with an inverse relationship with all kinds of cardiovascular events. That is as solid a finding as you can get in observational epidemiology. It is a very reliable prospective marker. It’s natural that the pharmaceutical industry and those of us who are interested in risk reduction would focus on HDL-C as a target. That is where the controversies come in.

Dr Nissen: It has been difficult. My view is that the trials that have attempted to modulate HDL-C or the drugs they used have been flawed. Although the results have not been promising, the jury is yet out. Torcetrapib, the cholesteryl ester transfer protein (CETP) inhibitor developed by Pfizer, had anoff-target toxicity.^[2] Niacin is not very effective, and there are a lot of downsides to the drug. That has been an issue, but people are still working on this. We have done some studies. We did our ApoA-1 Milano infusion study^[3]about a decade ago, which showed very promising results with respect to shrinking plaques in coronary arteries. I remain open to the possibility that the right drug in the right trial will work.

Dr Libby: What do you do with the genetic data that have come out in the past couple of years? Sekar Kathiresan masterminded and organized an enormous collaboration^[4] in which they looked, with contemporary genetics, at whether HDL had the genetic markers of being a causal risk factor. They came up empty-handed.

Dr Nissen: I am cautious about interpreting those data, like I am cautious about interpreting animal studies of atherosclerosis. We have both lived through this problem in which something works extremely well in animals but doesn’t work in humans, or it doesn’t work in animals but it works in humans. The genetic studies don’t seal the fate of HDL. I have an open mind about this. Drugs are complex. They work by complex mechanisms. It is my belief that what we have to do is test these hypotheses in well-designed clinical trials, which are rigorously performed with drugs that are clean—unlike torcetrapib—and don’t have off-target toxicities.

Elevated levels of high-sensitivity troponin T (hsTnT) and N-terminal pro-B-type natriuretic peptide (NT-proBNP) strongly predicted heart failure in patients with chronic kidney disease followed for a median of close to 6 years, researchers reported.

Compared with patients with the lowest blood levels of hsTnT, those with the highest had a nearly five-fold higher risk for developing heart failure and the risk was 10-fold higher in patients with the highest NT-proBNP levels compared with those with the lowest levels of the protein, researcher Nisha Bansal, MD, of the University of Washington in Seattle, and colleagues wrote online in the Journal of the American Society of Nephrology.

A separate study, published online in theJournal of the American Medical Association earlier in the week, also examined the comorbid conditions of heart and kidney disease, finding no benefit to the practice of treating cardiac surgery patients who developed acute kidney injury with infusions of the antihypertensive drug fenoldopam.

The study, reported by researcher Giovanni Landoni, MD, of the IRCCS San Raffaele Scientific Institute, Milan, Italy, and colleagues, was stopped early “for futility,” according to the authors, and the incidence of hypotension during drug infusion was significantly higher in patients infused with fenoldopam than placebo (26% vs. 15%; P=0.001).

Blood Markers Predict CKD Heart Failure

The study in patients with mild to moderate chronic kidney disease (CKD) was conducted to determine if blood markers could help identify patients at high risk for developing heart failure.

Heart failure is the most common cardiovascular complication among people with renal disease, occurring in about a quarter of CKD patients.

The two markers, hsTnT and NT-proBNP, are associated with overworked cardiac myocytes and have been shown to predict heart failure in the general population.

However, Bansal and colleagues noted, the markers have not been widely used in diagnosing heart failure among patients with CKD due to concerns that reduced renal excretion may raise levels of these markers, and therefore do not reflect an actual increase in heart muscle strain.

To better understand the importance of elevated concentrations of hsTnT and NT-proBNP in CKD patients, the researchers examined their association with incident heart failure events in 3,483 participants in the ongoing observational Chronic Renal Insufficiency Cohort (CRIC) study.

All participants were recruited from June 2003 to August 2008, and all were free of heart failure at baseline. The researchers used Cox regression to examine the association of baseline levels of hsTnT and NT-proBNP with incident heart failure after adjustment for demographic influences, traditional cardiovascular risk factors, makers of kidney disease, pertinent medication use, and mineral metabolism markers.

At baseline, hsTnT levels ranged from ≤5.0 to 378.7 pg/mL and NT-proBNP levels ranged from ≤5 to 35,000 pg/mL. Compared with patients who had undetectable hsTnT, those in the highest quartile (>26.5 ng/mL) had a significantly higher rate of heart failure (hazard ratio 4.77; 95% CI 2.49-9.14).

Compared with those in the lowest NT-proBNP quintile (<47.6 ng/mL), patients in the highest quintile (>433.0 ng/mL) experienced an almost 10-fold increase in heart failure risk (HR 9.57; 95% CI 4.40-20.83).

The researchers noted that these associations remained robust after adjustment for potential confounders and for the other biomarker, suggesting that while hsTnT and NT-proBNP are complementary, they may be indicative of distinct biological pathways for heart failure.

Even Modest Increases in NP-proBNP Linked to Heart Failure

The findings are consistent with an earlier analysis that included 8,000 patients with albuminuria in the Prevention of REnal and Vascular ENd-stage Disease (PREVEND) study, which showed that hsTnT was associated with incident cardiovascular events, even after adjustment for eGFR and severity of albuminuria.

“Among participants in the CRIC study, those with the highest quartile of detectable hsTnT had a twofold higher odds of left ventricular hypertrophy compared with those in the lowest quartile,” Bansal and colleagues wrote, adding that the findings were similar after excluding participants with any cardiovascular disease at baseline.

Even modest elevations in NT-proBNP were associated with significantly increased rates of heart failure, including in subgroups stratified by eGFR, proteinuria, and diabetic status.

“NT-proBNP regulates blood pressure and body fluid volume by its natriuretic and diuretic actions, arterial dilation, and inhibition of the renin-aldosterone-angiotensin system and increased levels of this marker likely reflect myocardial stress induced by subclinical changes in volume or pressure, even in persons without clinical disease,” the researchers wrote.

The researchers concluded that further studies are needed to develop and validate risk prediction tools for clinical heart failure in patients with CKD, and to determine the potential role of these two biomarkers in a heart failure risk prediction and prevention strategy.

Fenoldopam ‘Widely Promoted’ in AKI Cardiac Surgery Setting

The JAMA study examined whether the selective dopamine receptor D agonist fenoldopam mesylate can reduce the need for dialysis in cardiac surgery patients who develop acute kidney injury (AKI).

Fenoldopam induces vasodilation of the renal, mesenteric, peripheral, and coronary arteries, and, unlike dopamine, it has no significant affinity for D₂ receptors, meaning that it theoretically induces greater vasodilation in the renal medulla than in the cortex, the researchers wrote.

“Because of these hemodynamic effects, fenoldopam has been widely promoted for the prevention and therapy of AKI in the United States and many other countries with apparent favorable results in cardiac surgery and other settings,” Landoni and colleagues wrote.

The drug was approved in 1997 by the FDA for the indication of in-hospital, short-term management of severe hypertension. It has not been approved for renal indications, but is commonly used off-label in cardiac surgery patients who develop AKI.

Although a meta analysis of randomized trials, conducted by the researchers, indicated a reduction in the incidence and progression of AKI associated with the treatment, Landoni and colleagues wrote that the absence of a definitive trial “leaves clinicians uncertain as to whether fenoldopam should be prescribed after cardiac surgery to prevent deterioration in renal function.”

To address this uncertainty, the researchers conducted a prospective, randomized, parallel-group trial in 667 patients treated at 19 hospitals in Italy from March 2008 to April 2013.

All patients had been admitted to ICUs after cardiac surgery with early acute kidney injury (≥50% increase of serum creatinine level from baseline or low output of urine for ≥6 hours). A total of 338 received fenoldopam by continuous intravenous infusion for a total of 96 hours or until ICU discharge, while 329 patients received saline infusions.

The primary end point was the rate of renal replacement therapy, and secondary end points included mortality (intensive care unit and 30-day mortality) and the rate of hypotension during study drug infusion.

Study Showed No Benefit, Was Stopped Early

Part 5: Reality of the Diagnostics Industry in the US: Post Human Genome sequencing

5.1  Illimina and Roche

When Illumina Buys Roche: The Dawning Of The Era Of Diagnostics Dominance

Robert J. Easton, Alain J. Gilbert, Olivier Lesueur, Rachel Laing, and Mark Ratner
http://PharmaMedtechBI.com    | IN VIVO: The Business & Medicine Report Jul/Aug 2014; 32(7).

• With current technology and resources, a well-funded IVD company can create and pursue a strategy of information gathering and informatics application to create medical knowledge, enabling it to assume the risk and manage certain segments of patients
• We see the first step in the process as the emergence of new specialty therapy companies coming from an IVD legacy, most likely focused in cancer, infection, or critical care

When Illumina Inc. acquired the regulatory consulting firm Myraqa, a specialist in in vitro diagnostics (IVD), in July, the press release announcement characterized the deal as one that would bolster illumina’s in-house capabilities for clinical readiness and help prepare for its next growth phase in regulated markets. That’s not surprising given the US Food and Drug Administration’s (FDA) approval a year and a half ago of its MiSeq next-generation sequencer for clinical use. But the deal could also suggest illumina is beginning to move along the path toward taking on clinical risk – that is, eventually

• advising physicians and patients, which would mean facing regulators directly

Such a move – by illumina, another life sciences tools firm, or an information specialist from the high-tech universe – is inevitable given

• the emerging power of diagnostics and traditional health care players’ reluctance to themselves take on such risk.

Alternatively, we believe that a well-funded diagnostics company could establish this position. either way, such a champion would establish dominion over and earn higher valuation than less-aggressive players who

• only supply compartmentalized drug and device solutions.

Diagnostics companies have long been dogged by a fundamental issue:

1. they are viewed and valued more along the lines of a commodity business than as firms that deliver a unique product or service
2. diagnostics companies are in position to do just that today because they are now advantaged by having access to more data points.
3. if they were to cobble together the right capabilities, diagnostics companies would have the ability to turn information into true medical knowledge

Example: PathGEN PathChip

nucleic-acid-based platform detects 296 viruses, bacteria, fungi & parasites

http://ow.ly/d/2GvQhttp://ow.ly/DSORV

This puts the diagnostics player in an unfamiliar realm where it can ask the question of what value they offer compared with a therapeutic. The key is that diagnostics can now offer unique information and potentially unique tools to capture that information. In order to do so, it has to create information from the data it generates, and then to supply that knowledge to users who will value and act on that knowledge. Complex genomic tests, as much as physical examination, may be the first meaningful touch point for physicians’ classification of disease.

Even if lab tests are more expensive, it is a cheaper means for deciding what to do first for a patient than the trial and error of prescribing medication without adequate information. Information is gaining in value as the amount of treatment data available on genomically characterizable subpopulations increases. In such a circumstance
it is the ability to perform that advisory function that will add tremendous value above what any test provides, the leverage of being able to apply a proprietary diagnostics platform – and importantly, the data it generates. It is the ability to perform that advisory function that will add tremendous value above what any test provides.

Integrated Diagnostics Inc. and Biodesix Inc. with mass spectrometry has the tools for unraveling disease processes, and numerous players are quite visibly in or are getting into the business of providing medical knowledge and clinical decision support in pursuit of a huge payout for those who actually solve important disease mysteries. Of course one has to ask whether MS/MS is sufficient for the assigned task, and also whether the technology is ready for the kind of workload experienced in a clinical service compared to a research vehicle.  My impression (as a reviewer) is that it is not now the time to take this seriously.

Roche has not realized its intent with Ventana: failing to deliver on the promise of boosting Roche’s pipeline, which was a significant factor in the high price Roche paid. The combined company was to be “uniquely positioned to further expand Ventana’s business globally and together develop more cost-efficient, differentiated, and targeted medicines.  On the other hand,  Biodesix decided to use Veristrat to look back and analyze important trial data to try to ascertain which patients would benefit from ficlatuzumab (subset). The predictive effect for the otherwise unimpressive trial results was observed in both progression-free survival and overall survival endpoints, and encouraged the companies to conduct a proof-of-concept study of ficlatuzumab in combination with Tarceva in advanced Non Small Cell Lung Cancer Patients (NSCLC) selected using the Veristrat test.

A second phase of IVD evolution will be far more challenging to pharma, when the most accomplished companies begin to assemble and integrate much broader data
sets, thereby gaining knowledge sufficient to actually manage patients and dictate therapy, including drug selection. No individual physician has or will have access to all of this information on thousands of patients, combined with the informatics to tease out from trillions of data points the optimal personalized medical approach. When the IVD-origin knowledge integrator amasses enough data and understanding to guide therapy decisions in large categories, particularly drug choices, it will become more valuable than any of the drug suppliers.

This is an apparent reversal of fortune. The pharmaceutical industry has been considered the valued provider, while the IVD manufacturer has been the low valued cousin. Now, it is by an ability to make kore accurate the drug administration that the IVD company can control the drug bill, to the detriment of drug developers, by finding algorithms that generate equal-to-innovative-drug outcomes using generics for most of the patients, thereby limiting the margins of drug suppliers and the upsides for new drug discovery/development.

It is here that there appears to be a misunderstanding of the whole picture of the development of the healthcare industry.  The pharmaceutical industry had a high value added only insofar it could replace market leaders for treatment before or at the time of patent expiration, which largely depended either introducing a new class of drug, or by relieving the current drug in its class of undesired toxicities or “side effects”.  Otherwise, the drug armamentarium was time limited to the expiration date. In other words, the value was dependent on a window of no competition.  In addition, as the regulation of healthcare costs were tightening under managed care, the introduction of new products that were deemed to be only marginally better, could be substitued by “off-patent” drug products.

The other misunderstanding is related to the IVD sector.  Laboratory tests in the 1950’s were manual, and they could be done by “technicians” who might not have completed a specialized training in clinical laboratory sciences.  The first sign of progress was the introduction of continuous flow chemistry, with a sampling probe, tubing to bring the reacting reagents into a photocell, and the timing of the reaction controlled by a coiled glass tubing before introducing the colored product into a uv-visible photometer.  In perhaps a decade, the Technicon SMA 12 and 6 instruments were introduced that could do up to 18 tests from a single sample.

5.2  My Cancer Genome from Vanderbilt University: Matching Tumor Mutations to Therapies & Clinical Trials

Reporter: Aviva Lev-Ari, PhD, RN
http://pharmaceuticalintelligence.com/2014/11/05/my-cancer-genome-from-vanderbilt-university-matching-tumor-mutations-to-therapies-clinical-trials/

GenomOncology and Vanderbilt-Ingram Cancer Center (VICC) today announced a partnership for the exclusive commercial development of a decision support tool based on My Cancer Genome™, an online precision cancer medicine knowledge resource for physicians, patients, caregivers and researchers.

Through this collaboration, GenomOncology and VICC will enhance My Cancer Genome through the development of a new genomics content management tool. The MyCancerGenome.org website will remain free and open to the public. In addition, GenomOncology will develop a decision support tool based on My Cancer Genome™ data that will enable automated interpretation of mutations in the genome of a patient’s tumor, providing actionable results in hours versus days.

Vanderbilt-Ingram Cancer Center (VICC) launched My Cancer Genome™ in January 2011 as an integral part of their Personalized Cancer Medicine Initiative that helps physicians and researchers track the latest developments in precision cancer medicine and connect with clinical research trials. This web-based information tool is designed to quickly educate clinicians on the rapidly expanding list of genetic mutations that impact cancers and enable the research of treatment options based on specific mutations. For more information on My Cancer Genome™visit www.mycancergenome.org/about/what-is-my-cancer-genome.

Therapies based on the specific genetic alterations that underlie a patient’s cancer not only result in better outcomes but often have less adverse reactions

Up front fee

Nominal fee covers installation support, configuring the Workbench to your specification, designing and developing custom report(s) and training your team.

Per sample fee

GenomOncology is paid on signed-out clinical reports. This philosophy aligns GenomOncology with your Laboratory as we are incentivized to offer world-class support and solutions to differentiate your clinical NGS program. There is no annual license fee.

5.3 Clinical Trial Services: Foundation Medicine & EmergingMed to Partner

Reporter: Aviva Lev-Ari, PhD, RN
http://pharmaceuticalintelligence.com/2014/11/03/clinical-trial-services-foundation-medicine-emergingmed-to-partner/

Foundation Medicine and EmergingMed said today that they will partner to offer clinical trial navigation services for health care providers and their patients who have received one of Foundation Medicine’s tumor genomic profiling tests.

The firms will provide concierge services to help physicians

• identify appropriate clinical trials for patients
• based on the results of FoundationOne or FoundationOne Heme.

“By providing clinical trial navigation services, we aim to facilitate

• timely and accurate clinical trial information and enrollment support services for physicians and patients,
• enabling greater access to treatment options based on the unique genomic profile of a patient’s cancer

Currently, there are over 800 candidate therapies that target genomic alterations in clinical trials,

• but “patients and physicians must identify and act on relevant options
• when the patient’s clinical profile is aligned with the often short enrollment window for each trial.

These investigational therapies are an opportunity to engage patients with cancer whose cancer has progressed or returned following standard treatment in a most favorable second option after relapse.  The new service is unique in notifying when new clinical trials emerge that match a patient’s genomic and clinical profile.

Google signs on to Foundation Medicine cancer Dx by offering tests to employees

By Emily Wasserman

Diagnostics luminary Foundation Medicine ($FMI) is generating some upward momentum, fueled by growing revenues and the success of its clinical tests. Tech giant Google ($GOOG) has taken note and is signing onto the company’s cancer diagnostics by offering them to employees.

Foundation Medicine CEO Michael Pellini said during the company’s Q3 earnings call that Google will start covering its DNA tests for employees and their family members suffering from cancer as part of its health benefits portfolio, Reuters reports.

Both sides stand to benefit from the deal, as Google looks to keep a leg up on Silicon Valley competitors and Foundation Medicine expands its cancer diagnostics platform. Last month, Apple ($AAPL) and Facebook ($FB) announced that they would begin covering the cost of egg freezing for female employees. A diagnostics partnership and attractive health benefits could work wonders for Google’s employee retention rates and bottom line.

In the meantime, Cambridge, MA-based Foundation Medicine is charging full speed ahead with its cancer diagnostics platform after filing for an IPO in September 2013. The company chalked up 6,428 clinical tests during Q3 2014, an eye-popping 149% increase year over year, and brought in total revenue for the quarter of $16.4 million–a 100% leap from last year. Foundation Medicine credits the promising numbers in part to new diagnostic partnerships and extended coverage for its tests.

In January, the company teamed up with Novartis ($NVS) to help the drugmaker evaluate potential candidates for its cancer therapies. In April, Foundation Medicine announced that it would develop a companion diagnostic test for a Clovis Oncology ($CLVS) drug under development to treat patients with ovarian cancer, building on an ongoing collaboration between the two companies.

Foundation Medicine also has its sights set on China’s growing diagnostics market, inking a deal in October with WuXi PharmaTech ($WX) that allows the company to perform lab testing for its FoundationOne assay at WuXi’s Shanghai-based Genome Center.

Foundation Medicine teams with MD Anderson for new trial of cancer Dx