AACC urges support for lab roles in precision medicine
Interview by Steve Halasey
Clinical laboratorians have long recognized the importance of laboratory medicine for guiding decisions related to patient care. But that recognition hasn’t always been conveyed to or understood by healthcare policymakers and payors, who have sometimes seemed indifferent to the pleas of the laboratory community for greater support.
In June, the American Association for Clinical Chemistry (AACC) gave policymakers and payors a persuasive new explanation of why it’s vital that laboratory medicine isn’t shortchanged.1 With much of the federal healthcare establishment now working to accomplish new goals in precision medicine announced by President Obama, AACC seized the opportunity to highlight the critical importance of clinical lab testing as a foundation for the development of this field. Summarizing its position statement on “Advancing Personalized/Precision Medicine,” the association wrote:
AACC strongly supports efforts within the healthcare community to personalize patient care and improve individual patient outcomes. Laboratory medicine plays a vital role in enabling the practice of precision medicine, including the identification of drugs or interventions that are best suited for each individual patient. We urge policymakers and payers to adopt policies that ensure access to and appropriate reimbursement for laboratory testing that will facilitate personalized therapeutic decisions.
No doubt, this statement will be among many topics on the minds of attendees later this month, when AACC will gather in Atlanta for its annual meeting and clinical lab expo. In the meantime, to find out more about AACC’s position statement, CLP spoke with a member of the committee that formulated the document, Roland Valdes Jr, PhD, DABCC, FACB, professor of pathology and laboratory medicine and of biochemistry and molecular biology at the University of Louisville School of Medicine, and chief of clinical chemistry and toxicology at the University Hospital, University of Louisville.
CLP: Are the terms “personalized medicine” and “precision medicine” synonymous, or do they suggest different aspects of the same trends in healthcare? How is each of those trends distinctive?
Roland Valdes, PhD: Precision medicine is about accuracy of diagnosis and efficacy of treatment. That is, the precision with which a physician can make a diagnosis and perform a treatment based on understanding the fundamental mechanism of the disease. Personalized medicine is about formulating a treatment plan based on an individual’s specific characteristics—usually meaning a person’s genetic characteristics—as opposed to those of the general population. That is, the treatment is targeted to a single, specific individual.
Personalized medicine has actually been practiced for many years. What’s changed now is the accuracy with which it can be practiced. Precision medicine is really about completing the vision of personalized medicine.
In practice, these trends complement and enable one another. Consequently, some people use the two terms interchangeably. But they are also often parsed as denoting the distinctive meanings I’ve described. In either case, both the trends and the terms that refer to them do overlap and reinforce one another.
CLP: The promise of personalized medicine was foreseen even before initiation of the Human Genome Project in 1990. Why is it taking so long to come to fruition?
Valdes: The Human Genome Project was a general exercise in building knowledge, independent of any specific application. How that knowledge comes to be used depends on the applications desired in a particular field, such as medicine. A key driver of current advances in the field is the availability of technologies that are making possible real-time medical applications of that knowledge.
Importantly, applications of genomic knowledge in medical settings are being made possible, in general, through the efforts of clinical laboratorians and clinical laboratories and, more specifically, through their development of novel test approaches. This is the reason that pathology and laboratory medicine are so important in this particular space.
The time-lag between 1990 and now is not actually very surprising. Most progress involving new medical knowledge has been historically slow to be adopted, and clinical uptake of new products and protocols usually trails education by several years. But frankly, in some areas related to genomic testing, development and adoption are actually moving pretty quickly—even when compared to other advances that have been considered disruptive innovations in medicine.
CLP: As clinicians and pharmaceutical manufacturers began developing targeted therapeutics, was it generally a surprise that there might also be a need for companion diagnostics to help guide the use of such products?
Valdes: From the perspective of Big Pharma, I think, personalized medicine was initially only about market segmentation. As opposed to the industry’s previous “one pill fits all” marketing concept, personalized medicine was seen as a distinctive way of approaching the marketing of products that could not be used in all populations.
But as early as 1997, we published a paper that pointed out the importance of laboratory medicine becoming a driver in personalized medicine.2 That paper led to the 2001 spin-out of our firm, PGXL Laboratories, from the University of Louisville, and opened the door to a lot of activity in the personalized medicine space. Together, the paper and lab helped to spark a revolution in generalizing the application of pharmacogenetic testing to patient care.
Even so, it wasn’t until 2010 that the National Academy of Clinical Biochemistry published the first guidelines for pharmacogenetic testing in clinical laboratories.3 Since then, thankfully, some very good guidelines have been written, and several of them are excellent for encouraging pharmacogenetics researchers to engage with clinical laboratorians.
Nevertheless, for many years pharma companies maintained that this kind of engagement wasn’t needed. I give many lectures around the country, and I have often heard pharma reps say, “we really don’t need this kind of genetic information to help identify drug interactions.”
But some firms eventually realized the advantages of working with clinical laboratorians to produce companion diagnostics, and now circumstances are forcing this issue upon them. When pharma companies are developing novel, targeted therapeutics, for example, FDA is asking that they also develop and provide companion diagnostics suitable for stratifying the patient population for which the drug is appropriate. It all really makes sense; it’s just a process of evolution that, I think, pharma companies have now acknowledged.
In his book The Innovator’s Prescription, Harvard’s Clayton Christensen wrote that, basically, diagnostics will trump therapeutics.4 I believe this, and I think we’re slowly seeing it happen.
CLP: Now that the outlines of personalized and precision medicine are becoming clear, what are the key roles that require the performance of clinical laboratory testing?
Valdes: The criteria are the same as for any other type of testing that laboratories provide, including accuracy, precision, predictive value, and so on. But special circumstances can create a need for unique approaches to testing.
One special consideration that we know about is the need for harmonization among test results. Across all the sites and types of testing that are performed, it is essential that we have well-defined quality assurance and quality control parameters.
Most of the testing we associate with personalized medicine is linked to genetics and DNA testing, which is made possible by novel technologies and novel techniques. Over the past few years, laboratory medicine professionals have contributed a great deal to the development of such novel tests. Referred to as “laboratory-developed tests” (LDTs), they have become a central means for advancing personalized medicine.
Why now? Frankly, because novel technologies and sophisticated techniques have become available at rocket speed—much faster than anything else I’ve witnessed in my long career in medicine.
CLP: Clinical lab testing contributes to the diagnosis of patient disease, the selection of therapies, ongoing monitoring of therapy, and so on. How would you gauge the current status of testing for those purposes? Are they all equally well understood, adopted, and paid for? Are some types of testing more advanced than others in this regard?
Valdes: In a nutshell, no: they are not well understood, not well adopted, and clearly not well reimbursed. There are two major problems that we face today: clinical acceptance and reimbursement.
Although economic analyses have provided a rational basis for anticipating that personalized medicine testing will result in overall cost savings, a lot of skepticism still remains among payors. And with every new test, there is a question about which payor will be the first to take a chance at applying this kind of testing in order to reduce costs. At PGXL Laboratories, we have some very progressive clients that are now testing and taking advantage of this potential savings. From the overall perspective of cost savings and the advanced provision of healthcare, I think the results are promising.
But the biggest problem we face, except with a few progressive organizations, is reimbursement. The Centers for Medicare and Medicaid Services (CMS) has turned out to be surprisingly receptive to the advantages of personalized medicine diagnostics, and the agency is now paying for some of the new pharmacogenetic tests—not all of them, but some. These favorable decisions on their part certainly represent a recognition of the potential value of such testing.
But when it comes to reimbursement, we still have a long way to go—particularly with private payors.
CLP: It is a rarity to hear somebody link CMS with a progressive approach toward clinical laboratory testing.
Valdes: It is surprising to me also, because for many years I have been concerned about CMS not paying for tests, or reducing reimbursement rates for this and that test. But the recently created Center for Medicare and Medicaid Innovation is being funded to the tune of $1 billion a year for the next 10 years for the purpose of developing innovative ways of providing healthcare. And in some respects, I think that CMS is working hard to establish a new policy direction. I’m not going to say that CMS is perfect, by any means. Certainly, the agency needs to better utilize its coverage with evidence development program. But I think our next challenge is going to be with private payors.
CLP: Why has it been so difficult to communicate to healthcare stakeholders the importance of laboratory testing in the context of personalized or precision medicine?
Valdes: Innovation and disruption begins with doubts about the status quo. In the context of personalized and precision medicine, it all begins with new understandings about the genetic mechanisms of action and the fundamental bases of a biomarker’s functions. If one believes in fundamental pharmacology and its basic principles, it’s difficult to argue that understanding an individual’s metabolic or receptor status is not important for selecting or dosing medication. But not all providers understand this. So education remains paramount—and it’s really within our grasp based on the technology that’s available now.
For example, one of the biggest hurdles in laboratory medicine comes from the fact that healthcare providers often do not know what to do with the information a laboratory provides. We solved that challenge by having labs provide reports that include direct guidance to its “actionable information.” Those are two words that I think will be very, very important moving forward in this discipline.
At PGXL, for example, we provide a product called the “PGXL Primer,” which is a report that guides the physician step by step on how to use our pharmacogenetic information for each patient they’ve tested. That’s a way that we and others have closed the education gap and overcome the obstacles in the way of introducing personalized treatments to individuals.
CLP: AACC’s recent position statement makes a number of specific recommendations for Congress, FDA, the payor community, and the laboratory community. Does the organization see any of these groups as posing especially strong obstacles to the furtherance of lab testing for personalized or precision medicine?
Valdes: Although I am a member of the committee that helped to compile some of these recommendations, I can only speak for myself on this. Policies that restrict innovation by imposing undue regulatory burdens or establishing unachievably high proof-of-principle requirements are really problematic.
For example, FDA and Congress are considering imposing new regulations on the development of laboratory-developed tests, which include many of the most promising tests related to personalized medicines. But it will be really counterproductive if labs are required to undergo FDA accreditation for all the changes made to a lab test every time that test is performed. If FDA decides to regulate LDTs the same way that it regulates medical devices and diagnostics—which is what the agency is proposing to do—that will pose a tremendous problem for just about all clinical laboratory testing. It would really be a setback. So we need to reach some kind of balance and compromise in this area.
CLP: Where do you think that balance should lie? How should LDTs be regulated?
Valdes: Again, speaking for myself, I think we need to get documentation about the LDTs that are being performed. But frankly, LDTs are already regulated through the Clinical Laboratory Improvement Amendments of 1988 (CLIA). Those regulations are pretty specific about the things a lab must do in order to provide a test that is acceptable for medical practice. Laboratories are subject to inspection, their procedures are carefully defined and controlled, and they are required to indicate the clinical utility of each LDT. I believe that continuing with CLIA regulation, but perhaps strengthening it a little bit, will be much more effective than turning this area over to FDA—especially if FDA is going to use device regulations as the model for how to manage and clear LDTs.
Currently, when a lab receives an order for a laboratory-developed test, it puts together the necessary reagents, performs the test, and provides a result linked to some clinical utility. Under the terms proposed by FDA’s draft guidance, the lab would not only need to seek agency approval or clearance for the base version of the test, it would also have to submit an application to the agency every time it made a slight change in that LDT—a common occurrence necessary for optimizing test results—and then await agency approval before finally performing the test.
It’s going to be very, very problematic if labs have to get FDA permission every time they make a slight change in an LDT. In some of the major reference labs, 90% of the tests performed are LDTs. It would result in enormous costs for any laboratory to take so many LDTs through the FDA review process.
CLP: The recent statement on precision and personalized medicine asked Congress to promote or fund the adoption of health information technology, the creation of health data repositories, and the performance of basic and translational research for new tests. What should the relative roles of government and the private sector be for these purposes?
Valdes: AACC believes Congress should provide funding and incentives, whereas the private sector needs to continue innovating in the area of health information technology in order to advance the science that informs and underlies personalized and precision medicine.
The advances taking place today in digital pathology, telepathology, and telemedicine are revolutionary. And all of these tie into the use of health information technologies. The idea is to be able to combine these specialized reports into a single cohesive picture that provides clinical guidance, and there’s no better way of accomplishing that goal than by promoting health information technologies.
One of the bigger problems we have found when tracking patients’ medications is the sheer number of prescriptions involved. The average number of medicines taken by people over 50 years of age is 12. And since many of those medicines are being prescribed by different physicians, no single physician has a handle on what the others have ordered, or why. This practice of polypharmacy is a seriously cost-ineffective way of managing patients and it has health-related implications.
But when it is coupled with new health information technologies, pharmacogenomics can make a huge difference in efforts to select the right medicine, for the right person, at the right time. Using health information technologies means that a person can carry their personal health information wherever they go, thereby improving the coordination of medical practice and reducing the enormous costs and health risks associated with a lack of coordination.
CLP: Does it concern you that venture capitalists are investing less than they used to in this area? Is the recent statement hoping to encourage Congress to offer further incentives for investments to advance this field?
Valdes: Yes. Most practitioners fundamentally believe that Congress should step on the gas a little bit more. It should provide funding and incentives for the development of new tests, and for streamlining the process in order to support even more innovation.
CLP: The position statement cautions FDA not to impose policies that might hinder test innovation. Can you offer specific examples of policies that might cross this line?
Valdes: The vast majority of personalized medicine tests are continuously being refined and improved as more specific information becomes available. If we impose current FDA regulatory structures on these tests, we’re going to hinder their development. The idea of regulating LDTs as devices is a good example of a place where innovation could be truncated at the wrong time, because the tests are still developing very quickly at this stage.
CLP: How important is it for the payor community to adjust its evidentiary requirements when making policy determinations for personalized or precision testing products?
Valdes: It’s very important that payors recognize the inherent limitations in datasets for tests that apply only to a subset of the population. With the limited funding that’s available, there’s only so much a lab can do to develop and conduct population studies on the massive scale needed to meet current requirements.
When contributing to the development of a new drug, labs can and do produce robust datasets, but it’s impossible to gather as much data as is typically required under current policies. One of the difficulties is that most clinical trials start out being fairly agnostic about whether study participants fall into the category of patients expected to derive benefit from the new drug, because if subjects do not respond to the drug they can be further selected as candidates for other medications that might be better suited for them.
In The Innovator’s Prescription, Christensen suggests that clinical trials will play an increasingly important role in drug discovery. As researchers investigate the decision tree describing which patients respond to the test medication and have a relevant biomarker, it becomes easier and easier to separate those who do from those who don’t. Those who do, become candidates for the drug being tested. Those who don’t, will have a different biomarker, and will become candidates for a different medicine. As researchers keep going down the tree, they little by little target medications to new, very specific targets. But at this point, the payor community is having tremendous difficulty with having to view such massive studies.
The first objective of private payor insurance companies is not to pay for things. Studies suggest that personalized and precision medicine testing can save money, but only some of those savings are realized in the near term. The rest is realized later—sometimes much later. And it this “later” group that causes difficulty for insurance companies, because they don’t always keep the same patient under their control for more than 4 or 5 years, and the benefits may arrive long after that.
CLP: Clinical laboratorians have advanced knowledge about how tests are performed, what limitations tests may have, and how tests should be interpreted in the context of an individual’s care. How important will these roles be for the future of the profession?
Valdes: Very important. Physicians want to provide the best possible patient care, and they’re going to become more and more cost-conscious as payment models shift. As clinical laboratory experts, we have knowledge to help them.
This knowledge will become even more in demand as the number of companion diagnostics and personalized medicine tests expands. There’s going to be more and more work in this area. Laboratory professionals have unique expertise that can help improve the quality of care within the confines of limited healthcare budgets. And, in fact, one of our roles in our own organizations is to help reduce the effects of budgetary constraints.
Laboratorians will need to be involved in educating and training the next generation of physicians. Along these lines, clinical chemistry training programs will be essential for the future of personalized and precision medicine. For the past 25 years, for example, my institution has offered a fellowship program that is approved by the Commission on Accreditation in Clinical Chemistry (COMACC). About 15 years ago, we reviewed what we believed to be the future of clinical chemistry and toxicology and identified three important new subspecialty areas: clinical pharmacogenetics, diagnostic proteomics, and diagnostic informatics. Realizing that these fields represent the future of medicine and how professionals are going to interact with the concept of personalized medicine, we added subspecialty training in each of those fields to our clinical chemistry training program.
Those three subspecialties, I think, are going to drive the future of clinical laboratorians in their profession. And in addition, we’re also beginning to educate and train medical students on this concept of being more accurate and more precise when using diagnostic information coming from the laboratory.
CLP: For decades, in vitro diagnostics manufacturers and clinical laboratories have each felt the pressure of reduced reimbursements for their products and services. Can the promise of personalized or precision medicine be achieved in an environment where restricted revenues discourage investment in testing innovations?
Valdes: Yes, if allowed to try, I believe it can. In the US healthcare system, the payment paradigm is increasingly moving toward capitation. Precision medicine and personalized medicine can play important roles in improving patient care and reducing overall short-term and long-term healthcare costs by identifying appropriate tests that lead to better treatment options. Earlier diagnosis, for example, can improve outcomes and reduce the need for more health services.
I think we’re heading in the direction where recognition of personalized and precision medicine is going to make a difference in payor models for reimbursement. For a lot of the goals of personalized and precision medicine, I believe that this promise can be achieved.
Steve Halasey is chief editor of CLP.
- Advancing personalized/precision medicine. [Position statement.] Washington, DC: American Association for Clinical Chemistry, 2015. Available at: www.aacc.org/~/media/files/position-statements/advancement_of_personalized_medicine_position_statement.pdf?la=en. Accessed June 28, 2015.
- Linder MW, Prough RA, Valdes R Jr. Pharmacogenetics: a laboratory tool for optimizing therapeutic efficiency. Clin Chem. 1997;43(2):254–266.
- National Academy of Clinical Biochemistry. Laboratory analysis and application of pharmacogenetics to clinical practice. Valdes R Jr, Payne DA, Linder MW, eds. Laboratory Medicine Practice Guidelines. Washington, DC: American Association for Clinical Chemistry, 2010. Available at: www.aacc.org/~/media/practice-guidelines/pharmacogenetics/pgx_guidelines.pdf?la=en. Accessed June 28, 2015.
- Christensen C, Grossman JH, Hwang J. The innovator’s prescription: a disruptive solution for healthcare. New York City: McGraw-Hill, 2009.