The Panelists

Ginny Adams, RN, MPH, Coverys

Ms. Adams, Senior Risk Consultant at Coverys, has a clinical background in critical care and emergency nursing. She is trained in, and has facilitated organization-wide work in, both Lean and the Deming philosophy of improvement. She is committed to continually improving systems and processes to enhance patient safety, outcomes, and care.

Josiah Allen, OneOme

Mr. Allen, Director of Medical Affairs at OneOme, has been involved in clinical research, product development, and clinical adoption and implementation. An author of more than a dozen peer-reviewed publications in pharmacogenomics, he has lectured extensively on clinical applications of pharmacogenomics.

David R. Brown, MD, FACE, Children’s MN

Dr. Brown, a staff physician at Children’s Hospitals and Clinics of Minnesota and in practice at Pediatric Endocrinology and Metabolism, has served as a member of the clinical faculty at the University of Minnesota. He completed a fellowship in Pediatric Endocrinology and a National Institutes of Health post-doctoral research fellowship.

Kevin Halling, MD, PhD, Mayo Clinic

Dr. Halling is Co-Director of the Genomics Laboratory, DLMP, at Mayo Clinic in Rochester. A molecular pathologist and professor in Mayo Clinic’s Department of Laboratory Medicine, his primary area of interest is in the development of genetic tests that can be used for the diagnosis and treatment of sporadic and hereditary cancer.

Nancy J. Mendelsohn, MD, Children’s Minnesota

Dr. Mendelsohn is Director of Medical Genetics at Children’s Minnesota.  A board-certified medical geneticist, she has more than 50 peer-reviewed publications, book chapters, and presentations in top-tier journals. She is a recognized thought leader in lysosomal storage disorders and a Fellow of the American Academy of Pediatrics (AAP).

Alan Spiro, MD, Medica

Dr. Spiro, Senior Vice President and Chief Medical Officer at Medica, oversees Medica efforts to improve the quality of health care while managing costs. He emphasizes new models of collaboration to achieve those goals. His experience includes executive roles at health plans and health services businesses, as well as in clinical practice.

Melissa Truelson, MS, CGC, Ambry Genetics

Ms. Truelson, Certified Genetic Counselor at Ambry Genetics, specializes in variant assessment and result reporting as a cancer genetic counselor.She is a member of both the National Society of Genetic Counselors and the Minnesota Genetic Counselors Association, where she currently co-chairs the Public Policy Committee.

The Sponsors

Precision Medicine

A new approach to care

Minnesota Physician Publishing’s 47th Minnesota Health Care Roundtable focused on the topic of Precision Medicine: A new approach to care. Seven panelists and our moderator, Minnesota Physician Publisher Mike Starnes, met on April 20, 2017, to discuss this topic.

This afternoon we will examine the topic of precision medicine. Let’s start by defining some terms. What do we mean when we say precision medicine?

Dr. Brown: Let me start with three areas. Under what is known as P4 medicine, personalized refers to the individual medicine concept, which is predictive as we acquire large amounts of data. Acquiring this data allows us to have a preventive component for early intervention. The last component of P4 medicine is participatory—patients are now actively involved and contributing to the databases. Second, we have to look beyond the genome, and examine the entire sequence downstream from genomic expression. This should incorporate a “multiomic” consideration of genomics: transcriptomics, proteomics, metabolomics, epigenomics, the microbiome—and, even more important, the economics that will determine financing. Third, precision medicine will provide a scientific basis for health and wellness. We hope to find the transition zones between health and wellness and the onset of disease, allowing us, through lifestyle or therapeutic interventions, to introduce the preventive aspect and the underlying molecular perturbations that allow expression of the disease phenotype.

Dr. Spiro: I would add more “-omics.” Things like, “I have to pick up my kid at school-omic so I cannot go to my appointment,” or, “I have a spouse who is sick and I have to provide care-omic.” Precision medicine is personalized medicine from a science- and genomic-based starting point. For many people, their economics, their family concerns, and their spiritual beliefs trump any biology. Until we consider those factors, we will not be successful in what I will call individualized personalized medicine. Advances in pharmacogenomics may allow us to predict which medications will be most accurate, but some patients are not going to take the medication—they either cannot afford the copay or they just are afraid of it. We must consider social factors and spiritual factors, and start to see people as the complex, unique beings they are. As [Sir William] Osler said, “It is much more important to know what sort of a patient has a disease than what sort of a disease a patient has.” That takes more than biology.

Dr. Halling: I think you will hear that it is nothing new. We have been practicing precision medicine for 100 years. Take lung cancer as an example. In the beginning, we just referred to lung cancer, and then pathologists discovered that there was squamous cell and adenocarcinoma and small cell cancer, and that you could treat them differently. Now, with additional genetic information, we can further subdivide these and develop different therapies. So it is nothing new, it is just getting more and more precise with time.

What is the difference between pharmacogenomics and pharmacogenetics?

Mr. Allen: I believe they are largely synonymous. One aspect of pharmacology examines the clinical effects of drugs on various targets in the body. Pharmacogenetics looks at how your genes influence your response to medications. Genomics tends to have a broader view, thinking about gene pathways and interactions between genes. For the most part, though, they are pretty synonymous.

Ms. Truelson: Pharmacogenomics allows us to look at the tumor profile, the genomic profile in a given tumor, and choose a chemotherapy regimen that will effectively target that specific tumor. Not just the type of tumor, but that specific tumor in that specific individual.

When did we first start using the term “precision medicine”?

Dr. Brown: It dates to 2004, when the completion of the Human Genome Project allowed us to identify genetic variance in a meaningful way and to characterize individual traits. We differ by a very small degree, with perhaps 0.2 percent variability from human to human, but that was enough to define a concept of individuation. The digital revolution occurred concurrently, expanding our ability to accumulate large amounts of data. The establishment of social networks, and then the integration of the electronic medical record, really defined and made the terms “individualized medicine,” “personalized medicine,” and “precision medicine” realities. We could now incorporate and create a medicine that was proactive rather than reactive in the presence of disease.

What are some examples of how precision medicine is already in place?

Dr. Halling: I see it in oncology. In the case of lung cancer, Mayo has a panel that looks at quite a few different things simultaneously. Next-generation sequencing looks at 20-plus genes—things like EGFR, MET, RAS—and each of those help us make targeted therapy decisions. So if the patient’s tumor has an EGFR mutation, that predicts they will usually respond to specific drugs. In colon cancer, we routinely test for four different genes, and those predict resistance to EGFR therapy under certain medications. With molecular oncology, we are seeing 40–50 percent growth of test volume each year, because the targets that we know about increase each day and our understanding of the correlations with response to therapy grow each day.

Ms. Truelson: My background is also in oncology, but I know that there are other areas where this panel testing is becoming very important for therapeutic reasons. At Ambry [Genetics], our EpiRapid panel has 16 genes where there are therapeutic indications for epilepsy, and we have a similar panel in terms of cardiovascular disease. Neurology and cardiovascular disease are just some examples of where therapeutic indications can be managed based on test results.

Dr. Mendelsohn: We can now use genetic testing to identify children with complex disorders earlier, and the more rapidly we can identify their disease, the more appropriately we can treat them and know which direction to give families. It helps you direct the therapy if you know what else to be doing. For example, a child with Pompe disease, in addition to having cardiomyopathy, has myopathy. That guides the direction we should take with complex medical therapies, particularly for children in our ICUs.

Dr. Brown: The defined genetic variants that we can clinically correlate account for about 2.5 million nucleotide sequences—but remember, there are 3.5 billion base pairs throughout our DNA sequences in each of our single cells. New technologies address that with mass parallel screening, and one new system allows us to do a whole genome sequence in parallel in about 14 minutes, with commercial costs estimated to drop down to $500 within five years. As important as the genome is, though, I think that proteomics are going to be much more important. We have the ability to biosynthesize something on the order of 250,000 proteins, and proteins are able to do an amplification of the genome well over 1000- or 2000-fold in terms of genetic information.

Dr. Spiro: I want to put the economic lens on this. Five hundred dollars sounds cheap for a mass screening, but multiply that by every single person you will screen. From the insurance world I live in, you have now increased the cost of premiums by at least $500— probably much more, because you are also going to be dealing with follow-up care. You also have to consider pharmacologic technology. The more precise your targeting of drugs, the more opportunity for new patent medications, with average costs that may well exceed $1,000 or $10,000 per dose. It might be crass to talk about the cost issues, but they are real.

Dr. Mendelsohn: There are 67,339 tests now available that test for 4,974 disorders. Our job as clinicians is to make sure that we do the right test at the right time. In our institution, every single DNA-based test is reviewed by a genetic counselor to make sure that we are not overspending, but we can still save money if we get upstream of the diagnosis and do the right test. I believe it is premature to utilize proteomics clinically, but there are great gene tests and some good panels ready today that we can use for our patients.

Dr. Halling: Regarding this insurance issue, it’s my understanding that the testing expenses are a drop in the bucket compared to the pharmaceutical expenses. A lot of our testing presents an opportunity to avoid giving expensive drugs to patients who will not benefit from them, especially in cases involving immunotherapies and some targeted therapies. There needs to be a lot more work done on cost benefit analyses.

Mr. Allen: One new pharmacogenetic study involving 1,000 patients showed that 99 percent carried at least one actionable variant among five genes—so everybody in this room, basically, carries at least one actionable variant in one of five genes. When you consider the cost of testing and insurance, this has the potential to impact a large number of patients. We are going to have to think very strategically about how to implement this in a cost-effective manner and also to ensure that we help people avoid adverse events.

Dr. Brown: In the Pioneer 100 Project, 4,200-odd associations that were never appreciated before were identified in 108 randomly selected individuals. In 60 percent of those patients, early determinants of disease were identified that typical medical interventions did not interpret. Eighty-five percent of those diseases did not involve any pharmacologic intervention—they involved lifestyle and nutrition, so the cost of therapeutic intervention, from the preventive aspect, was economically reasonable.

What are the biggest challenges we face in making precision medicine part of everyday medicine in every clinic?

Dr. Halling: Consider the terms clinical validity and clinical utility. You could have a sound test, one that is sensitive and specific—one that is clinically valid—but when you use it in practice it may or may not change patient management, so it is just a waste of money. We need more studies, but it is a very high bar to pass, and frequently there are not enough patients at any one institution to accomplish it. Sometimes different institutions have to partner together. For genetic tests, they want to see evidence not only of analytic validity, but also of clinical validity and clinical utility.

Dr. Spiro: When I did training, I would teach people what I call the “secrets of medicine,” one of them being that 80 percent of illnesses get better on their own. As Voltaire said, “The art of medicine consists in amusing the patient while nature cures the disease.” Sometimes watchful waiting is the best thing for the patient—this is still a valid therapeutic and even diagnostic technique. We have this fundamental belief—a very American cultural belief—that all illness and death is optional. Maybe it will be someday. But right now, as we try to stave off all illness and death, we are making medicine unaffordable and unavailable to vast parts of the population.

Ms. Truelson: The world would be a better place if all human disease was understood, and the only way to do that is to collaborate and share data. There are lots of labs and research groups that hoard their data. Ambry has published several large papers recently, one in collaboration with Mayo that showed that some hereditary breast cancer genes may not confer the type of risk that we thought they did. None of that would be possible without collaboration.

Ms. Adams: Where does research stop and start and clinical care stop and start? We all judge care on the evidence-based best standard—what we accept as best care at that moment—and that standard is developing so fast. Somehow, we have to restructure and figure out how to use this knowledge to drive clinical care, to use outcome-based information to drive more research and more care. It’s a cycle, and collaboration is key.

Dr. Spiro: We need more clinical geneticists, physician geneticists, and genetic counselors, but the economics of it are lousy. The money in genetics is in lab testing and in potential genetic therapy. There is no money in the clinical practices. That’s just a fact, even though geneticists go through training that is akin to that of neurosurgeons. The representatives on this panel are making tremendous advances, but if you are an entrepreneurial company, you deal with investors and venture capitalists who want their 30 percent return each year. You need profitability or growth or both—and, again, that means high-priced labs.

How can we best update doctors and other members of care teams about advances in precision medicine?

Dr. Mendelsohn: We partner deeply with our clinicians so that, in each of our specialty areas, we have a particular physician who has an interest in genetics and genomics, and we have genetic counselors that we embed in our specialty clinics to partner with those clinicians. They interpret the tests and help empower the families with information. Equally important is the expertise of clinicians. Some practices, particularly at Sanford in their system in South Dakota, have done a great job integrating genetics in the primary care setting.

Let’s talk about genetic counselors. What do they do? What training do they have? How do they fit in?

Ms. Truelson: Most of us work in clinics alongside physicians, nurses, and other members of the health care team, meeting with patients and families. We gather family health histories to do a risk assessment for given hereditary diseases, and, if testing is done, we help interpret results for the patients and their families. Some genetic counselors, like me, work in industry or for laboratories, analyzing test results and assessing variants that we identify in people’s genes to determine if they are pathogenic or if they are harmless and issuing reports back to the physicians or genetic counselors who ordered them. Our training requires a two-year, accredited Masters program, followed by a board exam to be certified. Minnesota was the 22nd state to pass a genetic counselors licensure law. Starting in 2018, genetic counselors must be licensed in Minnesota.

Mr. Allen: When I interact with patients and clinicians, I get more questions on the drugs than I do on the genetics. We really see pharmacists as the people who will be the ultimate owners of pharmacogenomics. Pharmacy is, I believe, an underutilized resource on the health care team. These folks do four years of school entirely focused in pharmacology and pharmacotherapy, and then they count pills at Walgreens. I think we could do health care a great service by increasing their role, and pharmacogenetics is one of the things that they could ultimately own and bring to the table.

Dr. Mendelsohn: I could not do my job without a genetic counselor. When I came to Children’s, I asked that our genetic counselors be part of our professional staff. We now make sure that, even though they are embedded into the specialty clinics, they also come back to the genetics department to share information with colleagues and with our pharmacogenomicists. Having that team to share information and disseminate it into the frontline has been critical for success.

Dr. Halling: We probably have 30-plus genetic counselors at Mayo, at least 20 of them in the laboratory and maybe 10 or more in the clinic. Our head genetic counselor told me there are only 250–300 counselors graduating each year, even though there are 2–3 times that many positions available. Salaries for genetic counselors are higher with the laboratories, especially commercial labs, than they are in the hospital setting.

Ms. Truelson: The NSGC [National Society of Genetic Counselors] invested in a workforce study to address the shortage of genetic counselors, and training programs are looking at ways to graduate more students. When genetic counselors change jobs, the vast majority go from clinic to lab, not the other way around. When you look at professional status surveys, the reason for that is primarily job satisfaction relating to salary and—I’m going to get on my little soapbox here—administrative support. I know genetic counselors who are still addressing envelopes, getting their own records from other clinics, and spending time faxing things. There are some little things that can be done to support genetic counselors in the clinic and to keep them there.

Dr. Halling: We went live with whole exome sequencing about a year ago. Genetic counselors helped us interpret a lot of the data before it went to the lab director for final reporting, but it is hugely labor intensive. It can take a week to look at a case with one person, and that drives up costs.

How can we build reimbursement for genetic counseling into the system?

Dr. Spiro: There are many challenges, including the problems of getting genetic counseling and genetic expertise into smaller hospitals and smaller settings. Larger systems gobble up hospitals and clinical facilities all the time, and it becomes a contracting and business relationship—buying supplies is cheaper, and it is easier to contract with a managed care plan. We need the systems to step up and have true clinical integration of what they are doing. We need the managed care companies and the hospital systems to understand that, from a financial point of view, medicine is now team-based. When it comes to reimbursement, old-fashioned fee-for-service still rules the day, and that’s a major problem. Unless we recognize team-based pay appropriately, we will be behind the times in how we pay people, and our payment systems will get in the way of incorporating, not just genetic counselors, but pharmacists and other health care professionals in a meaningful way.

Dr. Mendelsohn: You can only do that if you are clinic-based. If you are a hospital-based clinic, you can’t. There are regulations around that, so we can’t bill for our counselors. And even when you do, which we have tried in the past, it is less reimbursement than the facility fee.

Mr. Allen: The problem is the same in the pharmacy world, where you have pharmacists who are not considered providers, so they are not able to prescribe or order testing. They have to do all of that under the supervision of a physician, and are not able to operate independently. There are certain medication therapy management codes that they are able to bill, but those do not pay well and so it is not financially feasible to integrate those types of pharmacists into the system. We need to find ways to make the value they bring match the reimbursement that comes back.

Dr. Spiro: This will sound offensive, but when you talk about reimbursement, it is a negotiation between large health systems and large payers. I would love to see the clinicians within the large health systems lobby to place more value on clinical services than on the facility fee. It’s a zero sum gain, obviously, but that’s what’s happening. Health systems are demanding certain prices for a facility fee and for other types of fees. For inpatient care we negotiate much more on DRG [diagnosis-related group] payments, but I’d love to see the health systems come to us and say, “We have to change our internal value system. We have to value the work of the clinicians much more, and we want our pricing to reflect that.”

Ms. Adams: You talked about payment as the reason we do not have these people on the front line working with our patients. I think that it is tied very closely with the whole culture of medicine. It is a physician-driven model, so you have to get your physicians to a point where they are willing to share that “captain of the ship” role with all these other people who are so critical to the patient’s care. It all has to evolve together.

This might be a good time to discuss the short-term future. Where will precision medicine be in three years, and how will it get there? How can we go from a place where we cannot afford to practice precision medicine to a place where we cannot afford not to do so?

Dr. Spiro: We have to get a whole different payment mechanism. There are lots of different ways to reimburse for health care. The first thing to understand is that we do not have a health insurance system in this country. Insurance is the idea of pooled dollars to cover the risk of unlikely events. Part of what we have is insurance, but mainly what we have is prepaid health care, and that’s different from insurance. There are certain societal decisions that have to be made in terms of how you spend that money for prepaid health care. Right now, we are dysfunctional in how we make those decisions.

How do we get precision medicine into the mainstream of day-to-day health care delivery?

Mr. Allen: I’ll give you one example involving clopidogrel, or Plavix, a blood thinner given to patients who have had heart attacks. In about 25 percent of the population, a genetic variant predisposes that medication not to be effective. In those patients, there is a 50 percent greater chance of heart attack or stroke than in patients who do not carry the genetic variant, and some of them go back to the hospital within 6–12 months with another heart attack, simply because the medication we prescribed was not effective. We could have known that going into it and prescribed them other medications that would have been efficacious. We also need to consider reimbursement. When I’m out talking to providers, the first question is, “What is pharmacogenomics?” and the second question is, “Is it covered?”

Dr. Brown: There is another dimension that contributes to cost effectiveness. Drug development and pharmacotherapeutics are partly predicated on outdated rodent models. The vast majority of drug development/drug testing goes on in murine models (mice genomics), which have little or no homogeneous relationship to the human genome. It’s a poor model for drug development and therapeutics. At Recombinetics, we attempt to create more realistic disease models by inserting human genes into larger animal models that are much closer to us, both physiologically and genetically. Most of you know that if you look at the 10 or 12 most currently prescribed drugs, it is hard to find efficacy greater than 20 percent in many cases. We are dealing with a flawed biology that underlies our whole pharmacotherapeutic science. I am currently engaged in creating a much more valid biology that underlies drug development and gets drugs to market that have efficacies in the 80–90 percent range and have safety profiles that approximate those same percentages.

Dr. Mendelsohn: In precision medicine and in anything involving genetics, we have to remember the importance of privacy. If you evaluate one family member, you may indirectly be evaluating another family member. If you have a first cousin to a child and you do a genetic test and it is positive and they both have the same abnormality, you have just diagnosed both of their parents without doing a single test.

What does a patient need to know about precision medicine?

Mr. Allen: First, genetics does not mean everything and it is not deterministic. It may be because of the movie “Gattaca,” but there’s this societal thought that genetics is 100 percent deterministic and that my genes control my destiny. If we can dispel that myth today, we will have done a huge service. Second, there obviously are big concerns about privacy, as well as fears and misconceptions.

Ms. Adams: Patients should know this is available. I live in the rural part of the state, and these sorts of things are not front and center. We need to raise awareness of what is available, as well as its limitations. As we just discussed, genetics is not deterministic, it is one piece, along with lifestyle choices and your environment. Patients should be aware, before testing even starts, of the privacy implications, the unforeseen information that may be identified, and all sorts of other things. They need to know the possible impact on their medical care, not only medically, but financially, emotionally, and socially. With all of medicine, we have been trying to bring the voice of the patient to the table as we develop things in our hospitals, clinics, and communities. We must not forget, as we move forward, that the patient is a key contributor.

Ms. Truelson: Not all tests are created equal. The $250 pharmacogenomics test that Josiah’s company can offer is very different than a $250 test that my company could offer. They just look at different things, and that is not always apparent to the general public. Even if you dig deep into some company’s website, you may not get the full picture of what is available.

Do we need precision medicine advocates?

Dr. Spiro: We do need patient advocates. People want help to get through a horribly difficult health care system. Also, the question of what patients should know starts with what doctors should know. Doctors should know what is important to the patient, and they often do not. For some patients, the diagnosis is not the most important thing, and for some patients, the biology is not the most important thing. Many studies show that, when you inform patients about statistical probabilities and the risk ratios of tests, it makes no difference in their decision-making. It takes a distant second to what their friend has told them. We have to understand what the patient is basing their decision on, and then the patient should know whatever else they need to know to make the best decision.

Dr. Brown: Patients do not only want to know what defective gene they have or what particular disease they could be facing. They want to know how well they are, how free they are of certain disease processes, and they want someone to explain the statistical certainty. We talk about medical care, but the term health care is even more relevant if these tools can define health in terms of both wellness and disease in a realistic and meaningful way. I am all for finding people to guide patients through this process. You need the expertise of specialty physicians to decide what tests to order, but once that process is complete, I believe we can have patient advocates guide that process, do the interpretation, and enhance compliance. It is one thing to make a diagnosis or to prescribe a therapy, but it is very, very difficult to maintain that compliance factor to optimacy. That is often a big factor when you consider the efficacy of drugs or of lifestyle interventions.

Dr. Mendelsohn: One of the difficulties is looking at how personalized medicine or genomics applies to large populations, and then applying that information to individual patients with complex disorders, particularly children. These are two very distinct populations. It’s important to stop and think about what a test is for and what population you are looking at. Looking at pharmacogenomics in adult populations or studying whether one drug is effective is important. But then if you look at pediatrics and you look at children, at the medically complex disorders in that small group of individuals, it actually is responsible for a huge amount of cost. That is where we can have a tremendous cost/benefit ratio and the ability to take those medically complex patients and funnel them through a genetics system.

Ms. Truelson: You just described a genetic counselor, a patient advocate who holds the patient’s hand through the process and helps them determine what is most important to them. I cannot tell you the number of times I walked into a room knowing what test I was going to order for a patient and then left not ordering anything because it was not what they wanted.

Dr. Mendelsohn: This is a very esoteric example, but we had a patient with CRIM-negative Pompe disease. Children with this disease have no nascent enzyme, and the body then attacks the enzyme, so when we give them exogenous enzymes or enzymes from outside the body, their bodies attack the enzyme. We developed a mechanism and a treatment for that particular child that we were then able to expand to other diseases. The opportunities to do that in a clinician’s lifetime are very few and far between. If we shared data and had the opportunity to look at pathways, we could learn a great deal.

Should there be increased consumer protection for over-the-counter products that claim quasi-medical validity related to genomics?

Mr. Allen: That’s a timely question. I believe that, without exception, every single one of the genes on one new panel marketed to consumers has an American College of Medical Genetics [ACMG] recommendation against testing for it. The questions that the FDA asked this company when they were seeking approval focused on whether any harm would come to patients, and did not really ask about the benefits of providing that information to patients. I was not particularly thrilled with how the FDA handled that particular incident. It is odd for me, being somebody in the laboratory industry, to be saying that, but I think that there do need to be protections in place. I have heard that there is a lot of snake oil out there, and the issue of clinical validity and clinical utility is really important. There needs to be some sort of oversight, some group that studies it and says whether these tests are helpful, not just whether they meet the bare minimum of not doing harm.

Ms. Truelson: Even the bare minimum of not doing harm might not be met. A group of genetic counselors at our company just presented an abstract at ACMG where we did follow-up testing on people who had had direct-to-consumer testing, and there was a 40 percent false-positive rate in genes like BRCA and TP53. BRCA genes increase risk for breast cancer and ovarian cancer, and TP53 causes a very, very severe hereditary cancer syndrome. So the false-positive rate is a little scary.

Dr. Spiro: There is also a consumer demand for insurance carriers to pay for some tests that may have no clinical validity. From an administrative expense point of view, the appeal processes are significant, which is fine because due process should be done, but clearly, if there were better mechanisms on the front end to define clinical utility in general—not just for genetic tests but for a lot of other services as well—that would be extremely helpful.

Dr. Halling: Both CLIA [Clinical Laboratory Improvement Amendments] and the FDA require IVD [in vitro diagnostic] manufacturers, if they are going to market a test, to determine its analytical performance, characteristics, accuracy, reproducibility, and analytical sensitivity. Before you can sell a test as a kit, for instance to other laboratories, you have to establish all that to get FDA approval and clearance, and then include it in the package insert. For laboratory-developed tests that are not FDA approved, and for reference laboratories, you should be able to visit their website and see what the analytical performance characteristics are. It is sort of a black box, because sometimes you do not know how the test performs, since they consider it proprietary. That does not even begin to address clinical validity/utility.

Dr. Brown: There is a serious deception in those direct-to-consumer testing services. As bad as the genomic testing may be, the companies are really looking at the value to themselves of accumulating data and gleaning data and then marketing it. They are actually selling a $99 test to you that is costing them $400 to conduct, but collectively they may be retrieving $1,000–$1,500 by selling the collected data anonymously.

Dr. Halling: There needs to be some assessment of the risk to the patients when they interpret the results by themselves. If it is too hard for them to understand by themselves, then you need a health care provider in between.

Dr. Spiro: Remember that these tests give risk profiles. They don’t give positives and negatives, and yet they are interpreted as positive and negative by many physicians. The patients themselves are also looking for answers, and we have to be careful of that. Also, in an era where electronic medical records are transferable and there may be breeches of security, we have significant privacy issues. Many people don’t pay attention to privacy when it comes to their own data, because they don’t really understand the downside risk. There is also a small minority who just want nothing to do with electronic records.

What can be done to prevent precision medicine from creating a new level of social disparity in health care?

Dr. Spiro: There is a tremendous risk of that, and it is already being seen. For several years, India has had a problem with selective abortions of female embryos because of the cultural bias towards male children. That is one example of genetic information—in this case, ultrasound information—having a social impact.

Dr. Mendelsohn: I’m more concerned about access to testing—whether people believe they can afford it, and whether they have access to insurance. One of the beauties of being in an organization that cares for all children and serves multicultural children is observing how different cultures accept children with differences. One of the great things I’ve learned from our Somali families and our Hmong families is that they are so accepting of kids with differences and kids with special needs. But I worry about them having access to the same kind of testing, particularly for those families that do not have insurance.

Mr. Allen: Virtually all new drugs are tested and approved in Caucasian populations. There is a lot of disparity in how pharmaceuticals work in people of different ancestral backgrounds, and pharmacogenomics provides an opportunity to understand the molecular and biological basis for that disparity and address it. For example, the immunosupression drug tacrolimus works one way in the Caucasian population and is much less effective at normal doses in African American populations, because they have a particular gene variant that makes them metabolize that drug much more rapidly. Pharmacogenomics can help us use that drug more effectively, because we can test for that gene and dose the medication appropriately.

Dr. Brown: In my lifetime, all new technologies—computers, compact discs, and large screen televisions—have been accompanied by social disparity. I remember being harangued years ago in a record store because I was buying CDs, which at that time were three times the price of analog records. I was called an elitist. All new technologies can create disparities, but technology related to precision medicine may lead to cost reductions, and not just because it will become more common. Technology will approach issues from multiple directions, from drug development and different strategies to training new health care professionals. All of this will reduce cost and limit liability, which will further reduce cost. I’m hopeful that this will be a democratizing process. We will look back in 25 years, when we have true, operational precision medicine, and see that this was smart and cost effective, and it has brought better health and better wellness to everyone.

Dr. Halling: Intuitively speaking, I totally agree, but there are very few studies that demonstrate reduced costs. You read about cases involving children who had workups for two, three, or four years without a diagnosis, and then they got exome sequencing and their doctors finally arrived at a diagnosis. If they had received that exome sequencing at the beginning, it would have saved a lot of money. There are other anecdotal examples of how you could use the technology to save money, but it’s not absolute proof.

Dr. Spiro: My rose-colored glasses are not quite as rosy as yours. Of course, I hope that this leads to better care and lower costs, but as I look around this panel, I don’t see any ethicists, I don’t see any sociologists, and I’m concerned. I’m in the insurance industry, where underwriting was once a major factor. Even though we’ve eliminated most of the medical underwriting, there is still a risk of using data like this to exclude people from services.

What are some of the most important considerations about ethics and professional liability as they pertain to precision medicine?

Ms. Truelson: One of the big ethical issues is how we obtain truly informed consent before testing even begins. With the myriad of tests that are out there, and the complexity of those tests, will patients really understand what they could get back from a test result, and how that result could impact their relatives? And whose responsibility is it to notify these relatives who may be at risk? Is it the physician? Is it the patient? That needs to be discussed beforehand.

Dr. Halling: I see this with oncology patients. If you’ve got a patient with advanced melanoma and you identify a BRAF mutation, which suggests that they will respond to vemurafenib, you give them the drug. Yes, the drug will eradicate the tumor for 6–12 months before it comes back, so you have bought a little time, but ultimately they are going to pass away unless immunotherapy works. You don’t want to withhold therapy from somebody who might benefit, but how much benefit has truly been derived? It comes back to what we can afford. Any one of us would want a chance.

Dr. Spiro: There is a real tension between individual ethics and population ethics. From a population ethics point of view, if you have just a set amount of resources in society, you could argue that no one over a certain age should have dialysis, for example. You could argue that six months of life under an extremely expensive protocol is not a societally ethical point of view, because that same money could be used for X-number of high risk maternity patients. From an individual ethics point of view, if I’m the patient, or if someone I love is the patient, I would want everything done. Insurance carriers and other payers are often left in the difficult position of being arbiters of what they will pay for. That is probably the wrong place to have it, but that’s where it sits today, or with government, which is also the wrong place.

Dr. Mendelsohn: There are also the ethical issues associated with deciding to test or not test, determing who we test and when we do so, and what we do if we do not test. It is important to me to make sure that we are doing things for children and not to children. If you have an answer that indicates a child may or may not survive, should you be doing the cardiac surgery or putting them on ECMO [extracorporeal membrane oxygenation]? What about when families disagree with a physician? For me, any good genomics program needs to have an ethicist involved.

Mr. Allen: We have a panel that covers a number of different disease states and so, if a psychiatrist orders our test, they will get additional information about genes that might affect drugs that are outside of their specialty. There are a lot of concerns about that.

In terms of professional liability, one of the benchmarks is always the definition of the standard of care. Does this new genomic precision medicine redefine the standard of care, and how will that impact potential malpractice cases?

Ms. Adams: If I bled to death after taking a certain drug, could you argue that it was because of that drug, and that my doctors could have known that would happen if they had tested my genes before prescribing it? Not today, because that testing isn’t accepted practice and we’re not doing it routinely, but at what point will it happen? Right now, we decide malpractice cases based on expert opinion and the best judgment of our physicians. Are we looking at a future where that will no longer be the case—where, instead, malpractice cases will be decided based on the data available about the condition or the drug at that moment in time? Lawsuits happen years after the actual event. In this rapidly evolving world, how do you look back five years and guess what was actually known and what was the standard of care at the time? Does it become incumbent on physicians to use, not so much their professional judgment and experience, but to pull up the data and treat the patient based on the data? And how do we ensure that every physician has the most current and the best data?

Mr. Allen: One very interesting case going before the Supreme Court concerns a patient whose testing identified a particular variant. At the time, they didn’t know the significance of that variant, although it was being studied by the company. Years later, they found that the variant was pathogenic, and that certain people in the company were doing research in that particular area but had not yet made a definitive call. This will be a very interesting court case, because it will raise questions about the responsibilities of the laboratories and the people who study and identify particular variants to report new findings.

Dr. Spiro: We often do not recognize that facts change, including facts involving genomics, proteomics, and other gene expression issues. I used to say that what I did professionally was to translate the black and white of insurance contracts into the shades of gray of medical practice. That’s not easy, but that’s what we’re really talking about. Lawyers will say there are shades of gray in law, but it is a lot less than it is in medical practice.

Life-saving and life-enhancing advances happen faster than they can be assimilated into the delivery system. What are the most important things that we must do to expedite this process?

Dr. Mendelsohn: Lots of laboratories today can sequence. Moving forward, the challenge will be to marry the data with the clinical phenotype and to interpret it with the software. Data will probably live in a secure cloud, to be pulled down into the electronic medical record. For pharmacogenomics, we could start with drug-drug interactions and drug-gene pairs. For clinical genetics, it could be early on, when we make definitive diagnoses. But moving forward, when should we reanalyze the data, and how will we pull that data in? What’s the right software to use? What’s the standard of care?

Mr. Allen: I think there is an ethical and professional responsibility for industry to provide education, provided it’s done the right way. There have been abuses of that system, so we need to build up trust again on both sides. There are often policies in place that make it difficult for industry to get in and talk to providers. One psychiatrist out in Columbus, Ohio, told me it’s really difficult to work with residents coming out of Ohio State, because they don’t know anything about the new drugs. They are 10 years behind, because they don’t let the reps in or the medical affairs folks in to provide any education about the new stuff that is coming out. Some of that is for legal reasons, but I think more is due to the policies within health care institutions. We have to rebuild some trust so that we can work on that together.

Can value-based reimbursement methodology play a factor in expediting this assimilation?

Dr. Spiro: It certainly can, but again, there is danger lurking behind each of these opportunities. In the past, attempts at different reimbursement strategies have had problems. Back when we capitated primary care, some people said it would be the future of medicine. The idea was that we would understand the financial risk and would be able to manage it, and then direct care toward the highest utility. Instead, individual practitioners often misinterpreted it as saying, “Oh my god, I only have $8 or $50 per member per month to treat this patient,” when that wasn’t the case. So we clearly need value-based pricing. We clearly need bundled payments of different types. Perhaps we also need capitation of some sort, but we need it with the right systems in place to make sure that it is being used responsibly and that it’s getting the correct results.

As we’ve seen today, precision medicine covers a lot of ground and offers both challenges and hope. What are the most important things we need to understand about precision medicine?

Mr. Allen: Precision medicine is not 100 percent and it’s not deterministic, but that doesn’t mean it’s not useful. It is available today and the costs are rapidly going down to make it more and more affordable. Obviously it comes with a lot of benefits, but it also comes with some challenges. We need to do a lot of work to increase education, both within the community and among providers, to understand how can we take precision medicine and harness it for the betterment of society and health care.

Dr. Mendelsohn: The potential is there for us to save lives, but we have to be judicious about when we do testing. We have to be careful about the promises we make and about our interpretations. Precision medicine is not a panacea. The potential to treat people and improve their disease state is there, but it’s not all ready for prime time. We have to be careful moving forward.

Dr. Spiro: Words matter. I’m actually uncomfortable with the term “precision medicine,” and that’s because so many parts of our definitions are still so imprecise. It worries me that we are calling it precision medicine. We put too much weight on it, and we leave out the sociologic and economic factors. Is precision medicine individualized medicine? Not really. It’s a component, and we have to understand how it fits in to everything else we do. We have to understand the ethical considerations. We have to understand how the payment systems will deal with this, when many of the labs and pharmaceutical companies are making tons of money off this and putting a burden on the entire system. Of course, they also contribute to the entire system. When you deal with certain pharmaceuticals and you can eliminate a lot of other hospital costs and surgery costs, from an economic point of view that’s great, but we have to figure it all out.

Ms. Truelson: Interpretation matters, and collaboration is extremely important. You may get one test done at a given point in time, and we know we can tell you one thing about the results, but that may change five years down the road. It’s important for people to realize that as we get smarter, how we interpret test results and what we do with them may change. Again, collaboration is key. We need to stop hoarding data. We need big laboratories and big research groups to share their information and collaborate so that we can give the best answer to the patients to effect positive outcomes.

Dr. Halling: It is a really exciting time. New discoveries are coming out at such a rapid pace, and it is very exciting, from a scientific point of view, to be involved with all of this. But there obviously are a lot of economic and ethical dilemmas that we will have to deal with over the coming decades.

Ms. Adams: I think that we need to educate patients. We need to make precision medicine known to people and to make it available. We need to provide patients with resources, and then we need to find ways to get them engaged and partnered with us, because everything you read says that it is more essential than ever for patients to be active partners in what we do. We’ve tried to make that happen in health care, and we’re making slow progress. It may be that we have to get very good in a hurry with engaging patients and bringing them around the table.

Dr. Brown: My colleagues have done an excellent job framing everything, so let me leave you with this. We’re going to make medicine great again.


Minnesota Physician Publishing Inc. © 2018